Germany experienced little increase in unemployment during the Great Recession, and has found “internal devaluation” relatively easy and pain-free, unlike most other countries. Why ? Some observations on the role of social capital in financial crises.

Theoretically, a drop in general demand in the economy would not necessarily have to result in unemployment, if businesses were willing or able to reduce their costs by cutting everyone’s pay, rather than sacking a part of the workforce. An alternative to a general pay cut would be to reduce everyone’s work hours, which still result in lower incomes for individual workers. But in the real world, neither scenario happens because:

• firms don’t want to demoralise their entire work force by cutting everyone’s pay, rather than quickly firing a large minority of the workforce they won’t ever see again ; and/or
• unions and/or labour regulations may constrain businesses from acting in this way, especially for lower-skilled workers ; and/or
• firms may want to keep the best workers and get rid of the worst, and a general pay cut will encourage the best to leave even in a recession.

Even if there is unemployment, wages do not fall quickly in response to the excess supply of workers (nominal wages are ‘sticky’) for many reasons, one of which is that a mismatch between old and new jobs may persist in the economy. It takes time for unemployed programmers and bankers to retrain as lawyers and master-carpenters, or adjust expectations downward and become cooks and taxi drivers. So waiting for wages to come down in order to clear the labour market or get rid of unemployment is a painful process. Government policies can shorten or prolong it, but one way or another it almost always takes longer than anyone wants.

The point is, if the adjustment in a recession could take place primarily through a proportionate reduction in everyone’s incomes, rather than through the unemployment of a large minority, the pain of recession could be widely shared rather than concentrated and the recession itself might also be shortened. But basically that ‘pain-sharing’ never happens, because it cannot. There’s a coordination failure in the economy as a whole.

( Note : Keynesians [the older generation kind] would dispute the preceding paragraph, but I preemptively wave away the objection. The subject here is “internal devaluation” in a highly open economy, and general nominal wage reduction in a single country is partial-equilibrium in the world economy at large, so even Keynesians should be okay with it. )

A useful simplification of the dilemmas facing an open economy in a recession is the Swann diagram (adapted from here) :

The intersection of the two curves is the sweet spot of internal and external balance, where both internal and external balances are zero. If you have high unemployment and a budget deficit, you need to borrow that fiscal shortfall from the bond market, or the IMF, or a generous benefactor. But if you can’t borrow or there are limits to borrowing, then you must cut back on spending or raise taxes (i.e., austerity) and make everything worse. A country which controls its own currency can devalue it and create a trade surplus, which can soften the blow of austerity (at least as long as your trade competitors don’t return the favour with a devaluation of its own).

But if you cannot conduct a sovereign monetary policy, then all your adjustment must be internal, i.e., as Andrew Mellon allegedly advised Herbert Hoover, “liquidate labor, liquidate stocks, liquidate farmers, liquidate real estate… it will purge the rottenness out of the system. High costs of living and high living will come down… Values will be adjusted, and enterprising people will pick up from less competent people…” Mellon was colourfully arguing for “internal devaluation”, i.e., essentially the same idea as a fire sale in a buyer’s market with an unsold inventory of cars or a glut of new houses.

But “internal devaluation” is prolonged and painful. The primary and fastest lever of alleviating recessions — monetary policy — is disabled by things like the gold standard of the 19th century, or the euro of today. The pain of adjustment-through-recession might be alleviated, even avoided, if everyone could “share the pain”, rather than a select group of people be unemployed. But that just does not happen, because it cannot — except in Germany.

§  §  §

It’s easy to forget now, but in the mid to late 1990s, Germany was derided along with France as one of the sick men of Europe. While the United States elated in economic triumphalism as its unemployment rate reached a historic low of 3.9%, the rates in the ‘big’ economies of Western Europe were mired in the 10-12% range, with Spain at even higher levels. And most of these economies were not in recession. As a cure for Eurosclerosis many argued for the implementation of Thatcher-like restructuring, with big tax cuts, big reductions in state spending, deregulation, and retrenchment of the welfare state.

In the case of Germany, following reunification it spent hundreds of billions of Deutschmarks to boost the eastern states — along the lines of 4% of West German GDP annually between 1991 and 1999. [source] This did not fully close the gulf between east and west, but German costs did rise, competitiveness was eroded, and Germany’s structural unemployment remained high.

Some time in the late 1990s, something changed within the German economy, and it was definitely a supply-side event, i.e., Germany’s labour costs began to fall relative to its trading partners’. Here’s an international comparison of unit labour costs, or wages relative to productivity [source], with 1995=100 :

For a high-wage country like Germany which had been reputed for labour rigidity, that’s a pretty steep rise in international competitiveness ! Germany accomplished an internal devaluation, which eventually paid off in a reduction of the unemployment rate :

as well as a big rise in Germany’s external surplus [in % of GDP, source : S. Wren-Lewis] :

These effects are often attributed to a series of reforms collectively known as the “Hartz plan” implemented between 2003 and 2005. They were a mixed bag of legislations enabling low-wage “mini-jobs“, curbing early retirement options and (most controversially, in Hartz IV) reducing long-term unemployment benefits. The Hartz reforms must have done something to reduce structural unemployment, especially for younger and older workers, if only by reducing unemployment benefits. That seems a reasonable interpretation, but it is in fact disputed by many people (e.g., 1, 2, 3).

Since at least half the improvement in German competitiveness predates those reforms by nearly 10 years, there must be other important reasons.

§  §  §

Even more interesting is that Germany’s recession in 2008-9 was quite shallow in terms of unemployment, i.e., the pain was “widely shared”. Although GDP contracted more in Germany than the USA or France,

the German unemployment rate barely blinked [both charts from here] :

But why ? In the 2008-9 recession, in response to falling orders, German firms reduced labour utilisation by cutting back work hours rather than by laying off workers. In fact, a whopping 90% of the drop in total labour inputs in the German economy in the Great Recession is to be explained by reduction of hours per worker ! This kind of “labour hoarding” does not normally happen during recessions in other countries, but did happen in Germany this time. This mini-miracle now has people on the centre-left all over the world seeking ways to replicate it in other countries, and has spawned a substantial literature.

In Germany, firms face restrictions on sacking workers in the first place, but in case of economic difficulties, they can apply to the Federal Employment Agency (Bundesagentur für Arbeit) for “short time” or Kurzarbeit. (Not the same as the 35-hour work week made famous by France, which is a statutory definition of full-time work.) The application entails a protocol for reducing work hours per worker, with the request triggering a process of consultations between the BA, management and works councils inside firms. After an agreement, workers can receive unemployment compensation in proportion to the loss in working hours. (There are great many more details, but they are really boring.)

There are sceptics who argue the effect of the Kurzarbeit system is exaggerated and the muted response of the German labour market to the Great Recession was due to the wage restraint of 1999-2008. Also, there had been a shortage of skilled labour which German manufacturers would not want to lose even in a recession. There’s some reason to think this is at least partially correct. Parts of the Kurzarbeit system go back as early as the late 19th century. In previous recessions, German firms had actually always resorted to hours-reduction more than in other countries. For example, during past recessions in the USA, ~30% of the reduction in labour inputs has been “intensive”, via adjustment of hours, whereas in Germany it has been ~60%. (source)

§  §  §

An explanation which best accounts for the dramatic rise in Germany’s competitiveness after the mid-1990s, that’s also consistent with its amazing labour market flexibility during the Great Recession, is contained in Dustmann et al. The argument hinges on a major institutional characteristic of Germany’s. Since it’s impossible to describe an industrial governance structure with charts or data, I simply quote from them [emphases mine]:

[ The Economist also had a really readable piece on a similar theme, an intimate portrait of how worker-management relations work inside a German company to maintain both quality and competitiveness. ]

Within a minimal institutional framework set up by statute, it is left up to workers and management to work things out themselves. Dustman et al.’s argument is equivalent to saying, Germany’s wage restraint [relative to productivity growth] since the 1990s was a natural, ‘market’ outcome of negotiations between management and labour in response to the erosion of competitiveness after German reunification. (Of course, it did help with the employers’ bargaining position with workers that part of their production could now be offshored to Eastern Europe after 1989.)

Most European countries have copied aspects of the “social partnership” model of ‘codetermination’ between government, business, and labour which in Germany is known as Betriebsverfassungsgesetz. But imitation has not produced similar results, and some countries have a more decentralised system of ‘co-determination’ than others. In France and some Mediterranean countries, the state frequently intervenes in the bargaining between business and labour, because (I would argue) there’s a much more adversarial relationship between labour and business. Contrast with the Anglosphere — it also has an adversarial relationship between business and labour, but then it has relatively few labour protections/regulations and basically no coordination of bargaining.

To simplify a little bit, the Anglosphere has social conflict and the state intervenes relatively little in the bargaining process; but in France and the Mediterranean, there is more social conflict and there is more state intervention. In Emmanuel Todd‘s scheme, the Anglosphere is individualistic and non-egalitarian, but France is individualistic and egalitarian. In the latter, the state must arbitrate and ‘fix’ things.

§  §  §

The relatively decentralised labour market institutions of Germany and the other northern countries reflect their lower level of social conflict. Germany, Austria, Switzerland, Denmark, Norway, Sweden, and Finland are well known for their efficient labour markets despite what, from an Anglo-Saxon perspective, should look like rigidities. And these same countries also do not have statutory minimum wage laws but still manage to have the de facto equivalent — wage floors set by bargaining between workers and management.

[The Netherlands, with its “Polder model“, also belongs in the list of decentralised bargaining countries, but it does have a minimum wage law. By the way, Germany has just recently passed a minimum wage law.]

Is it a coincidence that the northern Eurozone had a curiously parallel experience of the Great Recession (source) ?

Initially, the northern Eurozone countries experienced a similar, or deeper, crash in GDP than the southern countries ; yet their unemployment rates barely moved. [In the above, France is included in the “southern economies” even though it’s really intermediate between the two. So the southern economies actually look worse if France is excluded.]

So can this system be duplicated elsewhere ? I say no. In the Anglosphere it requires too many constraints on business than is currently regarded as politically and culturally palatable. In Europe outside the northern economies the constraints are regarded as too few ! The decentralised system of ‘co-determination’ is likely something that can only exist in societies with a lot of social capital, i.e., those which are cooperative, consensus-prone, and high-trust.

That social capital also shows up in the political system. A Swedish blogger who specialises in personality psychology explains how a “feminine personality” manifests itself in a “consensus democracy” :

Calling it ‘majoritarian’ actually understates the winner-take-all approach of the competitive model in the Anglosphere. Not a single government in the UK has received a majority of the popular vote since before the war, and the last time one came even within the upper 40s was in the 1950s. Yet, because of a combination of single-branch government and first-past-the-post voting, a plurality party in the UK can attain power and not necessarily achieve consensus in society even when implementing fairly radical legislation (e.g., Thatcher, who never got more than 42-43% of the vote).

In France, Thatcher would have been impossible simply because of the absolute-majority voting system. But even if a Thatcherite came to power with >50%, there would have been blood in the street from the minority out of power ! Of course the first few years of Thatcherism also provoked unrest in the UK, but one imagines it would be far worse in France with its long history of labour agitation and at-the-drop-of-a-hat demos.

§  §  §

Within the Eurozone, Germany’s current account surpluses have been a problem, which even the IMF and the US Treasury have criticised. But these surpluses were caused by internal devaluation — something almost nobody ever does voluntarily or accomplishes without pain. Germany’s labour market reforms, especially Hartz IV, are unpopular and, by German standards, pretty divisive. But they took place without massive social cost and without an extraordinary, compelling urgency, such as an external debt or a balance of payments crisis. I just can’t imagine something comparable taking place in France without triggering unrest even with a major emergency.

The ultimate cause of Germany’s problematic strength in the Eurozone is not an authoritarian culture, or primordial savings behaviour, or economic unilateralism, but its amazing institutional capacity and social capital. Surely that is as impressive as any feat of German engineering, or as exceptional as the logistical efficiency with which Germans of the past waged two-front wars.

An excerpt from Paul Fussell’s The Great War and Modern Memory.

The start of this month coincides with the centenary of a momentous calamity. There will be commemorated every possible consequence of that event, from the Bolshevik Revolution to the birth of the modern Middle East. They are all historiographic philistinisms !

Instead of contemplating great affairs of state, I heartily recommend the following as the best book on the calamity :

Instead of describing it in great detail, I quote some passages which I think give a flavour of the book.

It’s definitely not an “oral history”, but it is a book which explores the war through the diaries, the letters, and the poetry of an educated, cultured generation who had the first hints of “modernist” thought. But, unlike the actual modernists, they still wrote in high Edwardian style. At least a third of the book is quotations.

By the end of the book you get the distinct impression that the war fundamentally altered our rhetoric and modes of thinking. I don’t know if it’s true, but it certainly feels that way :

There’s also an amazing section on how Ruskin and Turner had influenced the way educated Edwardians looked upon sunrise and sunset, and how that influence interacted with morning and evening stand-to’s of the army in the trenches. And there’s also the brilliant reflexion on the “Uses of the Pastoral” in war imagery. But both are too long to quote.

The Great War and Modern Memory is unsentimental, mordant, and ironic in that subdued 18th century way. But by the time you finish the book, you will think to yourself, has there ever been such a moving book of literary criticism and history ?

On 1 August 2014, Germany declared war on Russia. Three days later, Britain followed suit with the following announcement :

“Owing to the summary rejection by the German Government of the request made by his Majesty’s Government for assurances that the neutrality of Belgium will be respected, his Majesty’s Ambassador to Berlin has received his passports, and his Majesty’s Government declared to the German Government that a state of war exists between Great Britain and Germany as from 11 p.m. on August 4, 1914.”

With such an inauspicious rhetorical beginning did Britain launch her Last Literary War.

Summary : New cultivars of cotton led to an unprecedented rise in the productivity of US southern cotton in the 60 years before the American Civil War. The Economist magazine may have said some stupid things about the subject in its review of Edward Baptist’s book on slavery, but it was fundamentally right to question the claim that antebellum efficiency gains were due to slaves being worked ever harder by their masters. (Note : I have a follow-up post, “Baptism by Blood Cotton“, which critiques in detail the central claims of Baptist’s book. There is also a brief note on slaves-as-wealth in Piketty & Slave-Wealth.)

I’ve said for a long time now that The Economist “newspaper” mostly produces glib, shallow hack pieces, despite somehow maintaining an august reputation to the contrary. Recently it published a review of The Half Has Never Been Told : Slavery and the Making of American Capitalism by historian Edward E. Baptist. In one sentence the reviewer used some vaguely injudicious language, and this caused outrage in the comments section and in social media. The Economist quickly “retracted” the review but quoted it in its entirety all the same. I reproduce the section which contains the economic-historical argument :

I’ve italicised the part which incited the outrage, but the final sentence is apparently the real howler in the eyes of the offended.

I’m more interested in the economics of the above claims. I haven’t read this just-published book, so I can only go by the review, and as far as I can tell neither Baptist nor the offended have disputed the reviewer’s characterisation of the book’s thesis.

The Economist‘s elf was lazily speculating, a priori, about what could have been the determinants of efficiency in southern cotton production. It’s possible the reviewer is familiar with some of the arguments and debates surrounding Time on the Cross : The Economics of American Slavery, one of whose many controversial arguments was that slaves were becoming ever more valuable property in the antebellum American South and were therefore better treated than commonly supposed. [Edit : Fogel restrictively argued, “The material (not psychological) conditions of slaves compared favourably with those free industrial workers”.] But I doubt that literature is known to the reviewer, because then he would have been familiar with recent research on the sources of increased efficiency of cotton agriculture in 1800-60.

According to the paper by Alan Olmstead and Paul Rhode, “Biological Innovation and Productivity Growth in the Antebellum Cotton Economy” [published version], the labour productivity of southern cotton grew 2.3% per year in the period 1800-60 :

The conclusions about the growth of labour productivity are based on the following data :

In short, cotton output grew significantly faster than labour inputs, i.e., slave labour productivity was rising quite steadily. The bottom of the table displays the market prices of cotton and slaves, which serve as a check on the productivity observation. Cotton prices were falling but a major input in the production process, slaves, was getting more expensive. This fact is strongly consistent with rising labour productivity.

Personally I can’t believe 2.3% per annum growth in productivity over 60 years could possibly be achieved by something other than major technological change. 2.3% per year is huge. To put that in perspective, consider that output per worker in English arable agriculture rose ~1% per year in the period 1800-1850. [See Table 15, Broadberry et al.] Or consider Olmstead & Rhode’s claim (page 35) that the mechanical reaper increased the productivity of wheat harvest by 50-100%. That is puny compared with the quantum leap in cotton picking productivity : 2.3% over 60 years amounts to almost 400%. How the hell does one achieve a quadrupling of output per worker just by increased effort alone, no matter how coercively obtained ???

Olmstead & Rhode attribute the productivity gain to the introduction of particular cultivars of the Mesoamerican cotton species G. hirsutum, more colloquially known as “upland cotton”. Its characteristics are well summarised by Stephen Yafa in Cotton: The Biography of a Revolutionary Fiber :

Yafa further notes that “green-seeded upland would become the one that blanketed the American South” ; and today “it accounts for about 95 percent of all the cotton grown and used around the world”.

Olmstead and Rhode devote several pages to the trial-and-error process by which various cultivars of G. hirsutum were carefully selected, hybridised (especially with a Mexican cultivar), and adapted to the soil and climactic conditions of the American south. (A more detailed literary account, full of observations by contemporaries on the importance of new cotton varieties, is found here.) They also document the failure of another promising cotton species G. barbadense, also known as “Sea Island cotton”, to take hold on account of its low yield.

The new Southern cultivars yielded more lint per boll and there were more bolls per plant. Not only that, some of the varieties also produced taller plants which were easier to pick. These photos show the dramatic difference that height could make (click to enlarge, source) :

For me the vastness of the efficiency gain alone speaks eloquently to some other factor besides ekeing more labour out of slaves until the pips squeaked. O & R nonetheless go through the motions and address the obvious alternate hypotheses for the increase in the cotton pick rate : (1) efficiencies of scale from the slave gang system ; (2) better management (including the possibility that slave owners are getting better at squeezing work out of slaves) ; and (3) the expansion of cotton production into western territories of the United States.

For the purposes of this blogpost the most important is #2 and O & R exploit a natural experiment in the daily picking rates of the high-yield Upland cotton cultivars versus the lower-yielding Sea Island cultivars :

Note that the Y-axis has got a log-scale ; the differences in the growth of pounds picked per worker-day of the two cultivars are colossal. In view of the above, it’s difficult to believe the euphemistic “managerial efficiency” was driving the secular rise in the pick rate of Upland cotton.

O & R also checked the improvements in yield on the same plantations over time before and after they switched from Sea Island to Upland. Unfortunately they produce no time series graphics for these observations, but they report large increases over time in the samples of plantations covered.

Finally, the botanical-genetic-technological explanation of the efficiency gains in US cotton production better accounts for the dramatic conquest of the world market for raw cotton by growers of the American south toward the mid-19th century. American cotton had 80% of the world share. Why didn’t this new American technology get diffused earlier and faster ? The Upland cultivar was adapted to the geoclimactic conditions of the American south and could not be rapidly adapted by the other primary cotton-growing regions of the world, such as India, Egypt or the Caribbean. Britain, the premier producer and exporter of cotton textiles, directly controlled many of these regions, but its manufacturers chose to import primarily American cotton.

So, YES, The Economist was quite fatuous and grossly ignorant in tendering the flaccid hypothesis that slaves might have been better treated, as a first-order counterargument to the claim that slaves were driven ever harder. But it was perfectly reasonable in doubting the latter claim in the first place.

At his blog the political scientist Chris Blattman rightly chides The Economist for not bothering to know the literature on the subject. (His post has been reblogged at the The Washington Post.) But then instead of getting to the heart of the matter — the efficiency of antebellum cotton agriculture — Blattman digresses about the literature on the efficacy and effects of coerced labour, with one link and graphic after another proving…. definitely not that an intensified effort at coercion of labour can quadruple worker output over six decades in a very labour-intensive production system.

On Twitter Blattman uttered to me the truism that “technology and (coerced) labour are complements”, as though no one has ever before decomposed the sources of economic growth into labour inputs, capital inputs, technological change, etc. The complementarity is beside the point. The point is that driving slaves harder, alone, could not possibly have accounted for the massive productivity gain in cotton.

PS — There were also some technological improvements in cotton agriculture unrelated to plant breeds that contributed to the efficiency of cotton production. Two most important are “scrapers” and machine seeders. But I do not stress those here. Also, the better soils of the more westerly lands in the “New South” were important.

EDIT : I’m waiting for Edward Baptist’s book to come out on Kindle, but Google Books has some excerpts. Unless there’s more to it than pp 127-8, it would appear the author uses the Olmstead productivity data, but merely hints at the botanical argument and dismisses that with a wave of the hand. He makes it sound (at least in those pages) as though it were only a matter of a single variety, Petit Gulf, of Upland cotton. But the plant breeding issue is definitely much deeper and broader. From another paper by Olmstead & Rhode (O&R 2007):

O&R 2007 also quotes from the plant breeding textbook, Breeding Field Crops, which indicates that the earliest cotton cultivars introduced to the American South,

O&R 2007 continues :

O&R 2007 goes into some detail on the plant-breeding aspect of the cotton cultivars. There’s definitely much more to it than just Petit Gulf.

The underlying claim in Edward Baptist’s “oral economic history” of slavery, The Half Has Never Been Told: Slavery and the Making of American Capitalism, is that slave owners, through the scientific “calibration” of torture, intensified the work of slaves in order to increase labour productivity by 400% on southern cotton plantations between 1800 and 1860. I argue the intensification claim is poorly supported, exaggerated, and misleading. This is a follow-up to my previous blogpost, “Plant breeding, not working slaves harder, drove cotton efficiency gains in the antebellum US South“. Warning : I’ve tried to make this post as readable as possible, but it goes into considerable detail, so it is dense and dreadfully dull !

(1)

In my previous blog entry, I argued that plant breeding drove cotton productivity gains in the antebellum US South, exploiting the work of economists Alan Olmstead and Paul Rhode, especially their paper “Biological Innovation and Productivity Growth in the Antebellum Cotton Economy” [the gated, published version is O&R 2008]. But having just read Baptist’s The Half Has Never Been Told,  I realise it really doesn’t matter whether the gains were achieved by new cotton cultivars, the invention of the cotton gin, or better soils in the western frontier of the “New South”.

The total output of American cotton increased by a certain amount in 1800-60, and the supply of slave labour devoted to cotton production increased by less than that amount. The difference implies a 400% increase in output per slave. No matter the cause of it, the four-fold increase in cotton per slave still had to get picked by someone.

Apart from land and labour, there were two limiting factors on the amount of cotton that could be grown in the first place. No matter how much cotton you could plant, it was of no use unless the “seed cotton” (the raw fiber) could be removed from the bolls of the plant ; and after the harvest, the seeds removed to produce lint. The gin alleviated the constraint on the speed of seed removal. But there was no mechanical cotton-picker until the 1930s, nothing comparable to the reaper for wheat. So the solution to the abundance of new frontier land and the potential for rapid seed removal was, to make slaves separate more fiber from the bolls faster.

The question is, which better captures what happened : (a) there was more acreage under cultivation (without necessarily better yield per acre) thanks to southwestern expansion, so there was more cotton to pick, and slaves were made to pick it faster ; OR (b) there was more yield per acre even with acreage expansion, so there was more cotton to pick, but slaves found it easier to pick it even if they also had to pick faster.

In my previous blogpost, I answered (b). If new cotton cultivars allowed slaves to draw more fiber out of more bolls on taller plants, then a lot of the work intensification was thanks to the plant.

(2)

Baptist attributes the higher pick rate to a revolution in slave-owner relations that separated “first slavery” from the second phase in the newly opened lands on the western frontiers of the old Southern states. Previously a system of passive resistance had prevailed where

In some regions of the Old South,

Even on tobacco plantations of the Old South, Baptist continues, surveillance was very often delegated to slaves “outside of direct white observation”, because the plots were small and scattered. However, with western expansion the “right hand” had figured out a profoundly new way to crush the passive resistance of the “left hand” and wring the slack out of the system :

“Full capacity utilisation”, as we might call it today, would be achieved by a fully white-surveilled and -directed system of gang labour, every bit as regimented as industrial factories. In this “pushing system”, a slave…

The primary incentives were negative, i.e., violence and torture :

Under the scientific principles of labour management, whipping was now no longer haphazard, let alone whimsical, but tightly linked to a quota-and-record system. “Enslavers used cotton-picking records to measure and record each enslaved person’s output. Such ledgers served, along with the scale and the whip, as key parts of the ‘whipping-machine’ system that raised cotton output steadily over time”. Therefore, it was not just the cotton gin, but this system of industrial management which allowed the revolution in cotton productivity.

(3)

When all the literary dressing is removed, Baptist’s argument can be summarised as: the “time and motion” system of whipping, quotas, and measurement drove each slave to economise on spare movement and pick cotton faster, faster, and faster. And the speed-up effect of this proto-Taylorism is what was responsible for the quadrupling of the daily pick rate between 1800 and 1860, from the average of 25 lbs (>11kg) per slave to 100 lbs. (>45kg).

Baptist, whilst dispensing with Olmstead & Rhode’s botanical explanation for the productivity growth, nonetheless relies crucially on their output estimates. In fact those data are the quantitative backbone of his thesis. Yet there is also another, unacknowledged debt. Whether he knows it or not, Baptist has used the work of Olmstead & Rhode to extend backward in time the efficiency argument advanced in a series of works on American slavery by Robert Fogel, the late economic historian & Nobel laureate, and his colleague Stanley Engerman.

Fogel & Engerman (F&E) were the authors of a controversial 1974 book, Time on the Cross: The Economics of American Slavery, which made a multitude of iconoclastic, revisionist arguments which were immediately challenged by pretty much everyone. (See their own 10-point summary of the book.) Economists in particular honed in on their claim that

In the above, by “efficiency” F&E meant not simple labour productivity, but total factor productivity (TFP), or the output per unit of all inputs into the production system, e.g., land, labour, capital, everything.

Economists descended like locusts upon the efficiency claim at a conference devoted to Time on the Cross, producing a series of critical articles later collected in book form ; F&E responded to the criticism in the The American Economic Review in 1977 ; which prompted 4 articles in counterresponse in the March 1979 issue of the AER  ; and to which F & E counter-counter-replied in 1980. In reaction to the numerous criticisms F&E corrected or refined the original claims and Fogel published a new book version of TOC called, Without Consent or Contract: The Rise and Fall of American Slavery, accompanied by 3 volumes of technical appendix. Finally, Fogel published a retrospective on that debate in 2003, which summarises the principal claims of WCC without much change. The history of this debate deserves a history of its own !

F&E’s claim of the productivity advantage of slave farms relative to free farms survived criticism and scrutiny. But Fogel had to shift his position on the causes of the productivity disparity. TOC had originally argued, the disparity was due to a combination of increasing returns to scale from plantation size AND both positive & negative incentives. ‘Positive’, meaning not just the lash but also rewards and ‘nice’ treatment as valuable property. Yet by the time of WCC, the “positive reinforcement” part was basically dropped. From WCC :

Here is the table of TFP comparisons worked out for 1860 (after correcting for some criticisms), reproduced from Fogel 1977 :

Also, Fogel toned down the crop-specific scale effects, and put more emphasis on scale economies in the supervision and management of a large number of coerced workers. This shows up in his reckoning of the source of the productivity advantage of large-scale slave agriculture :

(4)

But doesn’t the above support Baptist’s thesis ? Not exactly. Fogel found that in 1860 large slave plantations, on average, extracted 70-75% more work per hour out of their slaves than southern free farms, northern free farms, and small slave farms. I stress, this finding applied to 1860 only. What Baptist does is equivalent to turning this cross-sectional observation into a “time series” assertion.

We can reasonably surmise that slave cotton plantations of 1800 had rates of work extraction out of their slaves which were similar to the small slave farms of 1860. Therefore, the difference in “labour intensity” (actually labour utilisation), in 1860, between the large slave plantations and everybody else strongly suggests a ceiling on how much extra work effort per hour could have possibly been wrung out between 1800 and 1860 in the overall slave sector. Baptist may or may not realise this logical implication, but it is there.

Note : Fogel’s productivity estimates included all major southern crops, but cotton was the biggest as a share of the total and, since the efficiency advantage to gang slave farming was even bigger in sugar and tobacco according to Fogel, the 70-75% figure might be used, at a first approximation, for the “labour intensity” advantage in slave cotton. We could even just round it off to 100%. It still would not come anywhere to the 400% required by the implicit Baptist thesis.

(5)

70-100% can be proposed as the upper limit on the amount of “work intensity” that owners achieved with their slaves in the 60 years leading to 1860. But even that figure is highly uncertain. It is not known, for example, how many hours of work were devoted to cotton-picking by slaves, on average, before the 1850-60 period. As far as I know there is only one study on working hours on southern plantations, but it’s old, drawn from a very small sample, and does not specifically cover cotton. So we can’t match the evolution in hours worked per day per slave with the concomitant evolution in the pounds of cotton picked during harvest season. In other words, maybe there was intensification of work per hour, or maybe there was not and slaves simply worked more hours per day on cotton over time, especially if they worked on farms with a mix of crops.

Besides, does the change in the average cotton pick rate of 25 lbs/day/slave to 100 lbs/day/slave even represent such a major intensification of work effort, compared with free industrial factory labour ? First of all, the 1800 starting point of 25 lbs is low by any standard. Even the 1860 end point is not egregiously high. For example, The Economics of Mechanical Cotton Harvesting contains the following table of labour requirement :

Note the above records the average yield of lint, the product of ginning the “seed cotton” or the raw fiber in the boll which is what gets harvested. The seed cotton yield can be inferred from the lint yield with a ratio of 3:1 [Olmstead & Rhode 2010]. Thus for the southern regions in the years 1909-36, the average rate of hand-harvest is approximately 13 lbs per hour. In relation to slave faming, the daily pick rate for 10-12 hour days would be 130-156 lbs, which is consistent with Whatley 1987. The standard deviation is about 40 lbs. In other words, free labour in the early 20th century was hand-harvesting seed cotton at a rate about 1 standard deviation above the slave average for 1860.

Of course, the intrinsic biological yield in 1909-36 must have been higher, so the pick rate ‘adjusted’ to antebellum conditions should be lower. In The Half Has Never Been Told, this kind of adjustment is never made, in reverse, for pick rates. Nonetheless I agree with Baptist that, on average, free workers (or sharecroppers) were absolutely not willing to work as hard as slaves were made to work (a view supported by Ransom & Sutch). But my point here is not about the welfare costs of slavery, but only about how intense 100 lbs/day really was.

What about outliers with free labour ? “Cotton-picking contests” provide some clues. One such event in California supplies the following information for 1935 : “The average picked for the nine hours of work among the contestants who finished the contest was 647.17 pounds”. And the “average picked by the 72 contestants in 1937 was 645 pounds per person”, including a woman who picked 544 lbs. Even adjusting for higher intrinsic crop yield these rates are consistent with the information from the slave plantation records.

From another paper by Olmstead & Rhode, “Slave Productivity in Cotton Production by Gender, Age, Season, and Scale” :

These histograms display the distributions of daily pick rates for adult male & female slaves during the 1840-62 period. Notice the asymmetry; in statistics, these distributions are said to have “positive skew”, meaning that their right tails are fatter and/or longer than the left side, unlike the ideal symmetric curve of the normal distribution. According to O&R the mode (the most frequently occurring value) is 100. But it’s clear from just eye-balling the distributions that the high-performing outliers raised the mean.

Baptist touts the role of scientific management of slaves, which implies something else: the slave owners or their managers understood the strengths and weaknesses of their slaves, and operated according to the principle of comparative advantage. The “most able workers, those with an absolute advantage in picking, are assigned to other more difficult tasks where they possess an even greater advantage” [Olmstead & Rhode 2010]. This is remarked upon as well in Metzer 1975, Fogel 1977, and Toman 2006. I quote from Metzer’s “Rational Management, modern business practises, and economies of scale in the ante-bellum southern plantations” :

So, because the slave owners allocated labour according to comparative advantage, the best slaves (whether prime-age males, or the smartest of the population) were not all harvesting cotton. Those with lower picking ability were, in part, manning the harvest crew.

(6)

But there were also many slaves who picked well below the mean. Baptist makes considerable effort to describe slaves — by name, in most instances — who were made to pick 200, 300, even 400 lbs of cotton per day. From his narrative one might forget that 100 lbs/day was the actual average f0r slave cotton plantations in 1860. But at least he tells you the average. Here is the chart presented in The Half Has Never Been Told :

The above was adapted by Baptist from Olmstead & Rhode’s scatter plot :

The variation around the mean not only is pretty big but also it gets bigger with time. So when Baptist talks about all those farms with huge daily average outputs, be sure to remember there were many grossly unproductive ones.

(7)

Lastly, a short word on Caribbean sugar. It had achieved the apex of the plantation gang labour system in the 18th century. And the picture is less clouded by revolutionary technologies like the gin or vast frontier expansion or new cultivars. It was a crop production system particularly egregious for gang labour and cruel, high-mortality working conditions.

Eltis et al. 2005 generates estimates for labour productivity growth in 1674-1801 for Caribbean sugar by inferring from movements in the prices of sugar and slaves. The estimation method is not nearly as good as the one for southern slave cotton by Olmstead & Rhode. But for what it’s worth, under several different assumptions the long-run growth of slave labour productivity in Caribbean sugar was dramatically less than US cotton 1800-60 :

The above shows fluctuations of the level of productivity, not rates of growth. When productivity was rising, growth rates were on the order of 0.2-0.4% per year. But the movements are extremely volatile, with little or no long-term trend ultimately.

Postscript : In the long tradition started by Eric Williams, Baptist makes extravagant claims about the indispensability of the American cotton-slave machine to Western industrialisation. I guess that will be addressed in part 3 !

The global distribution of income in 1970-2006 in 20 charts. Not much text. Posting the charts because I thought they were neat…

In the late 1990s I used to follow inequality issues closely, but I’ve now mostly lost interest in the income distribution of the rich countries. After all, why care if the median income of an upper OECD country varies between 93rd and 93.25th percentile of the global distribution ?

But it’s more interesting to compare the income levels in absolute terms in different countries in relation with the global income distribution. The above chart, a global Lorenz Curve, divides the world population into ventiles or 5% blocks, and packs a lot of information into the simple format. That chart and the two following come from Branko Milanovic some of whose descriptions I paraphrase :

• In 2005, the poorest 5% of the US population were richer than the bottom 60% of the world population. The US income distribution spans 40 global percentiles, which are (relatively) few compared with….
• Brazil, a true microcosm of the world, contains the poorest 1% of the world as well as the richest 1%. (It’s not for nothing, Brazil’s reputation.) But it also has a pretty big middle class, by world standards.
• According to Milanovic, “…the Chinese top ventile attains almost the 80th percentile of the world’s income distribution. If we used percentiles, the top 1% of the Chinese would be better-off than 93% of the world population.”
• The average income of India’s richest ventile is only about as comparable as the USA’s bottom ventile. The top 1% of Indians “would not go past the global 80th percentile”.

For developing countries, global ventiles don’t convey fine-enough information to show how their very richest compare with the rest of the world. Yet some very poor countries do have people who are fabulously and opulently rich even by OECD standards. Also, the charts are all based on 2005 data. The patterns would probably look substantially different already in 2014. Unfortunately, household income information from around the world is not available regularly, so this lag can’t be helped.

With a different set countries also in 2005 :

Germany’s poor do extremely well by any standard… and the poorest third or so of Argentinians are poorer than the poorest third of Albanians ?! That can’t be a bias in favour of Albania ; Milanovic is a Serb…

Just for a look at African countries in perspective :

The three previous charts showed one of the most common ways to represent income distribution, the Lorenz Curve, which is a cumulative distribution function. Sometimes, however, it’s more interesting to look at income in terms of density distributions. Think of the bell curve of the normal probability distribution.

Unless specified otherwise, the following charts are taken from Maxim Pinkovskiy and Xavier Sala-i-Martin, “Parametric Estimations of the World Distribution of Income“.

In all cases, the X-axis shows incomes in 2000 international (PPP-adjusted) dollars, and the Y-axis displays population in absolute figures. The vertical lines indicate the standard incomes at $1 and $2 per day. The latter amount represents the UN-defined global poverty threshold.

Note : In some charts, either axis or both show values in log scale, depending on how big a range the numbers have.

Generally, wider the curve, the more unequal is the distribution of income. The taller the peak of the curve, more people are earning the income specified on the X-axis. If the entire curve is moving to the right over time, it means the population as a whole is getting richer.

Notice the world used to be almost tri-modal (three-peaked) but became slowly more bimodal, and now is almost uni-modal. Three separate worlds are blending slowly.

Here’s what it looks like when everyone’s income grows pretty rapidly — the peaks get taller, but the curves are narrower and more shifted to the right :

For China, there’s been progress over time but with a somewhat different pattern. Starting from a lower base the whole curves are moving to the right (everyone is making more money), but they get wider (incomes are more unequal).

I’ve argued before that progress in India started a little earlier than when the neoliberal reforms are conventionally dated (usually to 1991, but sometimes as early as 1985) :

Much more common in the world is progress that’s less dramatic, without income inequality changing all that much. Notice both Mexico and Brazil have very wide income distributions whose width hasn’t changed over time by very much. In the three following examples, the right half of the distribution has shifted right much more than the left half.

This chart comes from a different paper which uses standardised values for income, limiting comparability with the other charts here. But still the modest change in the African income distribution over time is evident :

Here’s what a really bad regression looks like — the right tail has shifted dramatically to the right but the left tail shifted just as dramatically in the opposite direction, and the centre of the curves hasn’t moved much to the right. The number of people earning close to the average income has fallen over time.

This shows how modest, in global terms, are the changes in the US income distribution. Of course the log-scale on the X-axis goes all the way up to $500,000 so the changes are not as modest as they look.

In general, people are too cheery when the economy is booming, and too gloomy about the future during recessions. Right now, pessimism about the consequences of “skill-biased technological change” is fashionable. But such pessimism should be regarded with the same scepticism as the never-ending predictions about the depletion of natural resources. The emptiness of human life and the infinity of human wants have always prevented mass unemployment when technological progress displaced workers. I believe that process will continue as long as there are some things machines can not do, albeit with a vast social difference from before.

(1)

Emil Kierkegaard posted a smooth and slick video which argues that a future is nigh in which supersophisticated robots substitute for unskilled and even moderately skilled human labour. I was captivated by the video until the end, when it very implausibly concluded that 45% unemployment might be in the offing.

In a blogpost entitled “The End of the Middle Class Era“, Razib Khan also appears to predict mass unemployment as a consequence of technological enhancements of productivity growth :

To be fair, I’m not altogether sure whether in the above he foresees mass unemployment, if only because he also references Tyler Cowen’s Average is Over. I found that book kind of vague (and the best parts of it were about chess). Those inclined to see his story as one of mass unemployment can point to the book’s suggestion that the employment-to-population ratio amongst prime-aged workers (25-54) will shrink over time. But Cowen himself waffles about his “zero marginal product” hypothesis, vacillating between the pessimistic “hopeless dregs of the earth” version and the less sensational ‘optimistic’ view.

In the latter, the “zero marginal product” workers are simply those who are the first to be unemployed at the start of a recession and the last to be hired during recoveries. In other words, these workers are only ZMP when general demand is modest but above zero when demand is brisk — such as when the US unemployment rate was at the historic low of 3.9% in 1999 or 2000. They also tend to show up as the primary victims of labour market rigidities. I think that’s the most sensible way to look at the Cowen scenario.

Unless machines can do everything and anything a human being can do, and those machines can be produced more cheaply than human beings can be employed, then I think “zero marginal product” is illusory.

Too many people assume that things like wages, incomes, and employment are nothing more than personality traits of workers, internal (innate and acquired) characteristics, rather than the outcomes of complicated interactions between individual characteristics and those of the rest of the world. Which is to say, too many people focus on the supply side of labour and neglect that there is a demand side.

(2)

In the past 250 years, technological progress has not caused unemployment because human wants have been infinite. Every time productivity (output per unit of input) rises, the implied extra income in the economy still gets spent on something (at least when there isn’t a recession), and extra work gets created to produce that something. In other words, fewer inputs may be used to make one unit of output, but more output always gets desired / created.

Environmentalists understand keenly that when energy prices fall, people frequently just drive more or fly more, or the savings get spent, ultimately, on something else that uses energy. Productivity growth produces the same effect. Which is why you never get mass unemployment.

(Won’t natural resources run out ? If you can believe robots can one day replace humans at everything, then it should be even easier to believe technology will also allow us to use fewer and fewer materials inputs per unit of output. [Or we will just mine the planets.] Materials productivity, better known as materials intensity of production, has also always kept improving and will keep improving. But that’s a different story.)

After the basic needs of food and shelter are satisfied, people go in search of other fulfillments — more caloric, varied, and exotic diets ; more living space to fill with ever more stuff ; 58 changes of clothes instead of 2 per year ; more leisure in the form of vacations and entertainment ; and ever more marginal extensions of life expectancy. That’s all very obvious.

But as people get wealthier, they demand not only more quantity of stuff, but also ever more trivial and even imaginary increments to the quality of goods and services. How else to explain the market for, say, honey in a jar that’s ‘raw’, unfiltered, unpasteurised, ‘fair-trade’, non-GMO, single-country-origin, single-bee-colony, and single-flower-species ?

Ironically, as production becomes more brutally efficient with labour-sparing technology, consumption becomes more ‘inefficient’. The hallmark of consumption by the rich has always been its labour-intensiveness. Think of aristocratic dining halls as recently as the Gilded Age, with one liveried footman for every guest at the long table in the dining hall.

That’s why ‘hand-made’ has snob appeal. Bespoke fetishists may think of it as “valuing timeless artisanal quality”, as does one London financial journalist who apparently has not only suits and shoes custom-made by hand, but also socks, neck ties, and (!) pocket squares. (When those silks stick out of the breast pocket, woe unto those rolled edges sewn with plebeian machine-neatness….) This tailor-blogger with a cult following makes suits by hand, or ‘deconstructs’ famous brands, and blogs about every lovely stitch. But in reality such sartorial epicureanism is about deriving more and more marginal utility out of sillier and sillier quality ‘improvements’. And such things point to the niche consumption fantasies of the merely upper-middle-class.

As long as there are still some things machines can’t do, I don’t see why that infinite-wants process can’t be extended indefinitely in the Cowtopian world where 15% earn a charmed living and 85% can at best aspire to the status of lumpenbourgeoisie. People — rich people — will just get even more petty, demanding, absurd, and elaborate in the infinity of their wants. I can’t imagine how exactly, but don’t underestimate the emptiness of human life !

The 15% of workers (in Cowen’s reckoning) who will succeed in the future must know how to work with AI-flavoured machines in a very complex production chain of human-machine complements. Such a profile would favour not only intelligence, but also conscientiousness, precision, discipline, and cooperative team work. The 85% who won’t make that cut, Cowen imagines, will scrounge around as petty entrepreneurs, freelancers, ‘consultants’, street-vendors, mobile taquería purveyors, and other genres of precariously ‘self-employed’. The lowest segment of that low segment will be “threshold earners” who proudly just “get by and who do not push ambitiously for a higher wage or stronger credentials at every step”. The culture will change to make that sort of thing hip, respectable, and freedom-enhancing. (Personally I think Cowen has perhaps dined at too many food trucks serving “Korean tacos” handled by tattooed hipsters and transacting with Paypal and iPad.)

And of course jobs should still exist for things like wiping after old and sick people, pet care, companionship services, hospitality, ‘aesthetic’ services, one-on-one tutoring, and city farmers catering to posh demand for heirloom ramps. Cowen also mentions jobs catering to the rich, but does not really take the idea very far.

Why can’t the conspicuous-luxury consumption sector account for a growing share of employment, as the top 10-15% capture a larger and larger share of GDP growth on account of skill-biased technological change (or what ever is the latest cause-du-jour of growing inequality)? Really, why not ?

(3)

With the advent of settled agriculture and the rise of cities, most human beings have most of the time worked for the rich in one form or another — whether as slaves/serfs, or sharecroppers, or free peasants paying confiscatory taxes, or factory workers, or service sector drones. There have been periods of freedom for frontier freeholders, but those don’t last very long.

But production is about creating wealth. My guess is, the future promises a world where the majority will still work for the rich, but less directly in helping to produce wealth, and more directly in enhancing the lifestyles of the rich. From adjutants in production to adjutants in consumption.

The size of the class that can afford absurdly infinite and infinitely absurd wants will expand. In 2013, the 99th percentile of household income in the USA was $385,000 and the average income of the top 1%, numbering approximately 1.7 million households, was $717,000. That’s nothing ! After taxes, after the big mortgage payment for the primary house and the vacation spare, after school fees for two offspring, there’s barely enough income left for a maid and a resident minder for the children. It’s definitely not enough to afford the whimsical last-minute day-trip by 90-minute suborbital flight from New York to Tokyo just to indulge in sushi at Sukiyabashi Jiro.

If economic growth of the future is severely skill-biased and concentrated in the top 10-15%, just think of the potential capacity for expansion in luxury consumption. The lifestyles of the top 0.1% will trickle gradually to aspirants in the 0.9%, just as the 5% will be closer to apeing the bottom of the top 1%, and down the line.

Given the number of workers employed “in service” at stately houses in England or robber-baron America, as late as 1914, or in developing countries today, you can easily imagine such “household establishments” being reproduced in the developed countries, but in more egalitarian-seeming ways. You’ll get the resident in-house staff ranging from manservants to henhouse keepers for each house owned in different locations. But a lot of the luxury service will come in the form of freelance labour that won’t seem nearly as bonded and mediaeval as hereditary footmen in a manor house or a fan-wallah in a maharajah palace. So there needn’t be a total reversion to the Downton Abbey world to suppose that more and more of the top 15% will consume snobbish labour-intensive goods as incomes grow and labour in the bottom 85% gets cheaper.

Why would a 5-percenter go to the chic supermarket or the picturesque farmer’s market for organic milk once a week, which is really for the sad people at the top of the fourth quintile, when he can keep a milk-cow of his own, hire a full-time farmhand to maintain it, and every morning partake of superfresh, hyperlocal, unpasteurised milk from his own arugula-fed, hand-massaged cow with the high-quality but low-yield magic-udder ? That doesn’t mean he won’t send his manservant (aka “personal assistant” in egalitarianese) to the market so he can also now and then try that subtly different, vaguely briney-creamy milk from the clover-fed Bronze-Age heritage-breed cows of the Vendée, delivered daily by the newest version of Concorde. Chic markets will continue to exist because the chic will always talk local and seasonal but will always want winter strawberries and asparagus from Chile and South Africa.

More 5- and 10-percenters will have full-time cooks, who won’t be inordinately skilled, but everything will be made from scratch from their urban gardens and small dedicated suburban livestock menagerie, or collected by the dedicated forager-hunter they co-employ with a neighbour. Right now the well-healed go to restaurants where “innovative” chefs work with ingredients harvested that very day, foraged that very day, caught that very day, and even slaughtered that very day. But the rich of the not-so-distant future could get a less elaborate version of that at home ! And why not, if it involves no effort on their part ? Of course they would still go to restaurants because of the comparative advantage offered by highly specialised chefs in originating more novel, trendy dishes.

I draw my hypotheticals from food because I know something about foodies and cuisines, and because in my unimaginative laziness I really can’t think of better examples. Social arrangements of the future are difficult to predict.

At least that’s the full-employment future I can imagine. And maybe I’m splitting hairs about the best flavour of pessimism ! But there is another version of the full-employment future.

But if human wants are not infinite ? Especially if machines can also replace humans at every task ? See Econ-Fi.

A quick note : Income inequality in pre-industrial societies was, in general, lower than in modern industrial societies, but traditional agrarian economies tended to be closer to their “maximum feasible inequality” than modern ones.

In “Russian Inequality on the Eve of Revolution” (I’m obliged by protocol to perform some kind of headgear ritual with Ben Southwood) there is a table which shows various measures of income inequality in pre-revolutionary Russia, along with some comparator societies :

It’s not surprising that income inequality should be low in agrarian, pre-industrial societies. You actually need some income to achieve substantial inequality ! But on second thought, that doesn’t properly put things into perspective, because you should compare actual inequality with the “maximum potential inequality” possible given the society’s level of per capita income. This point is vividly illustrated in Milanovic, Lindert & Williamson, “Pre-Industrial Income Inequality”, which supplies the interesting “inequality possibility frontier” curve :

The idea is, there is an absolute level of subsistence income for the mass of peasants who are creating value in the traditional society. Above that level is the surplus, or the amount of income which can be expropriated by elites ; and below that amount… starvation for the producers in the economy.

Thus, in a very poor society whose average income is close to the absolute survival level, the surplus extracted is pretty small and therefore inequality cannot be very high. But as average income rises, there is potentially more to extract. So the interesting question becomes, did pre-industrial societies at different levels of income have different “extraction levels” ? Put another way: did peasant incomes also rise when the average income rose or did the increase simply lead to more elite extraction ?

(You can click the above image to enlarge.) Subsistence income is assumed to be $300 in 1990 international dollars.

Incomes may have been more equally distributed in China in 1880 than in England in 1688, but that’s only because the average income was quite low in China. Perhaps more importantly, China in 1880 was closer to its maximum potential inequality than England in 1688. The chart below compares preindustrial societies with modern ones :

Some countries considered very unequal, like modern Brazil, are definitely more unequal than most pre-industrial societies, but it’s much farther from its “potential maximum inequality” precisely because it produces much more income than Peru in 1876. There’s just more to go around.

So for preindustrial or developing countries with very low incomes, it’s best to consider what Milanovic et al. call “inequality extration ratios” — the ratio of inequality to “maximum feasible inequality”, both measured in Gini terms. As average income rises with development, inequality also usually rises, but the extraction ratio should fall. Yet for some very poor countries today the “inequality extraction ratio” remains almost mediaeval :

Here’s also a full page of comparisons with modern and preindustrial societies.

Kenneth Pomeranz, in The Great Divergence, observed :

Mughal India does appear (in the table in the link above) to have been particularly extractive, but the other two historical values for India — especially the one for Bihar circa 1800 — not nearly as much. And you’d expect Bihar — one of the most benighted places on earth — to be a particularly exploitive place ! So Pomeranz may have been overgeneralising from the Mughal intensification of the preexisting zamindari system in north India.

A quick note on Piketty, slave-wealth, and US capitalism.

Matthew Yglesias had a Vox article thundering “American prosperity was built on slavery and torture” as part of his reaction to Edward Baptist’s book on slavery and American capitalism. Yglesias reproduced a chart from Piketty’s Capital in the Twenty-First Century to illustrate how valuable slaves were as capital :

To which Scott Sumner at Economics & Liberty responded by arguing “ending slavery made America richer” :

Sumner kind of buries his point, but he’s basically saying the chart plays accounting tricks. If you treated all labour resources in the same way, there wouldn’t be any (or much) change in the trajectory of American capital assets before and after emancipation.

But that’s the same thing Piketty says, except he said it an economically more compelling way ! A chart from earlier in Chapter 4 of his book doesn’t even mention slave-wealth :

Capital in the 20th Century :

Since labour, whether free or slave, receives flows of income, you could value labour just as you would a bond or an annuity or other assets with returns.

In fact, without the above reasoning, the irrelevance of emancipation to the economic growth of the United States would be a little puzzling in a micro sense. The wealth that had been held in the form slaves did not disappear into thin air, but was partially (*) transferred to the slaves themselves. Except nobody counts the present value of returns to human capital as “wealth”.

So ultimately, aren’t Piketty and Sumner saying the same thing ?

( * I say “partially” only because emancipated slaves did supply less labour than before.)

But still the slave-owners surely made a killing. Slave-cotton was certainly profitable year by year, but another question is, where did all the expropriated value of slave labour go ? Slave prices from Baptist :

You have to think of these prices as you might with those of tech stocks or housing prices. Just as high stock prices incorporate high expected returns, so the run-up in slaves prices incorporated expectations of profit from using slaves in production. People who owned slaves from early in the cotton era, or those few contrarians and “value-investors” who bought them when prices were low, must have made a killing. But investors and speculators who bought high might not have been able to amortise their investments.

Anyway, since both Southern raw cotton and British cotton textiles approximated perfectly competitive markets, the biggest beneficiaries of the value extracted from slave labour were almost certainly the consumers of British cotton textiles, i.e., tens of millions of people not just in Europe and North America but around the world. Even India had become a net importer of British textiles.

EDIT : A reader pointed me to this piece by Brad DeLong which makes a much more textbook case for the distribution of the expropriated value of slave labour amongst the pioneers in slave cotton as well as the world’s consumers of British cotton textiles.

From the plastics industry, a very concrete, real-life example of how conscientiousness matters in the use of a medium-level technology and how high(er)-technology might help.

In the past I have argued that technological progress can help developing countries overcome some of their deficiencies in human and social capital, and I instanced the social-infrastructure advantage of mobile telephony over landed communications. I never really fleshed that thought out, but there’s a more colourful version of the same idea here :

In this blogpost I illustrate how technology and conscientiousness might interact.

(1)

First I must briefly describe a medium-tech industry unadorned with the glamour of robotics or supersonic aviation. Most people barely give a second thought about them — plastic bottles. To be specific, bottles made of polyethylene terephthalate, or PET.

Today, around the world, plastic bottles made from this material are overwhelmingly dominant in water and carbonated soft drinks. But PET is also used to make containers for a wide variety of products, from liquid soap to pharmaceuticals to mayonnaise. You know what they are :

The hallmark of the PET material is that it allows for very strong but lightweight, soft, transparent, food-grade bottles (which can also be coloured, if necessary). A bottle of 35 grams, the weight of a pencil, is sufficient to contain 1.5L of carbonated soft drink without danger of explosion.

Before PET, glass bottles had been used for beverages, but obviously these were pretty heavy, easily broken, and very expensive to transport. And glass bottles require a very high fixed investment to manufacture, so scale was required to make production worthwhile. Other kinds of plastics (PVC, polypropylene, HDPE, LDPE, etc.) failed to pass muster on numerous grounds : too rigid, too brittle, not food-grade, too heavy, or opaque. Some of these, however, are still used for milk, motor oil, laundry detergents, etc.

But the PET material alone was not enough to do the trick. A new method of converting the plastic into containers was required — a two-step process called “injection stretch blow moulding”. In ISBM, the PET material is first melted and then moulded into a small, thick, rigid, intermediate proto-bottle with an already-finished neck, called a “preform”, which looks a bit like a test tube or perhaps a dildo. In the second step, these “dildoes” are then reheated, then stretched like a balloon, and only then blown into final shape with high-pressure air inside a steel or aluminium mould with the contours of the desired bottle. The result of this process was incredible strength combined with light weight.

Here is a 10-second animation visualising what the ISBM process looks like inside a bottle-making machine :

Here is also a 90-second video of an actual, live small-scale machine enacting the above process. The music is a little cheesy, but it really shows what’s happening inside the machine.

(2)

Because of all that, PET containers are ubiquitous around the world. And that’s due to the intensification of what economists call “multi-factorial productivity”. In order to deliver a single bottle to market, less material is required than ever before ; as well as less scale, less production space, less machine, less machine-time, less labour-time, less labour-attention, less energy, etc.

At first these machines required an operator with chemical engineering knowledge, because the plastic-melting and -shaping process was sensitive to ambient conditions like humidity and the parameters had to be frequently adjusted. But now PLCs can handle most of that.

The machinery industry quickly developed equipment for different levels of automation, speed, and scale. Thus a bottling factory could make 100,000 bottles per hour employing only a single operator with not much to do, running a giant automated machine like this :

But even a make-shift operation at the top of the Himalayas, using more labour, less PLC-automation, could produce a couple of hundred bottles per hour with a toy machine that could fit in a corner of your garage.

The high-end equipment are manufactured by French, Japanese, and German companies. At the lower end, there are — of course ! — hundreds of Chinese equipment manufacturers, almost all of them located in a single county of a single province, supplying bottle-making machines of various grades. Africa is full of “semi-automatic” Chinese machines.

In a way familiar to students of economic geography, the most highly developed countries tend to specialise in the manufacture of these machines, but developing countries are able to produce many of the accessories and downstream equipment in the bottling process.

(3)

There is a crucial, non-mechanical component in the bottle-making machine — the bottle mould. It’s “carved” from a block of steel or aluminium and then polished. The CNC machine which does the cutting is computerised, but bottle moulds have many different parts which must be finished separately, and there’s still a certain “artisanal” element which computers can’t yet duplicate (at least not in a commercially viable way).

So a lot of human attention, meticulousness, and precision are required in fashioning the mould. If those qualities are absent, a small mistake in the mould can result in unuseable bottles even if the machine is operating perfectly.

Many factories in developing countries import high-speed precision equipment with many sensitive moving parts from Europe or Japan. At the same time, they often try to save money on accessories by sourcing them locally. Bottle moulds are a frequent target for economising.

Here are two photos (click to enlarge) sent to me by an acquaintance, documenting a minor disaster at a bottling factory in a middle-income country which I promised not to name. (It has heavy processing industry, but is not known for manufacturing industrial equipment.) The bottling machine is a state-of-the-art Japanese model, but the factory owner tried to save about $300,000 (out of a total $3 million investment) by buying the moulds in his home country, rather than from the Japanese manufacturer.

The photo above shows the finished neck of the bottles produced by the Japanese machine but using a local bottle mould. The defective three on top have a thin, light film of plastic material around the circumference of the rim of the neck. Because of this film, the standard caps that are normally fitted onto the bottle in the bottling line, will not fit. The defects were caused by an overflow of molten material leaking out of a 0.1 mm gap in a region of the mould shaping the neck of the bottle. Click on the photo to the right.

This was not a design mistake, but carelessness in execution. It was not one-off, because several identical components had the same error. The mould-maker was perhaps a little busy pleasuring himself at the moment when the neck finish of the mould was being cut and shaped. You really have to look at the photos to know what I’m talking about. Or go to the slideshow at Google Photos. Now, I am not suggesting that this particular blunder by this particular supplier was necessarily due to some deep cultural or personality issue, although it might have been.

(4)

The minor disaster with the bottle mould is an example of how a little thing can make the whole or at least a big part of the whole useless. This idea was vividly and brilliantly illustrated with ice water and a $1 vise by the physicist Richard Feynmann. As part of the commission to investigate the explosion of the Space Shuttle Challenger, Feymann — at a news conference no less — argued cold made the rubber of a crucial O-ring used in the space shuttle inelastic, and this unleashed a chain of events which ultimately led to the disaster.

The “O-ring ring theory of economic development“, named after Feymann’s demonstration, posits that modern production requires interdependence and cooperation amongst workers performing tasks at different stages of production. The value created in such a process could be ruined in one stroke by just one non-performing link in the chain. The theory is usually expressed in terms of skill or collaboration, but it might as well also be in terms of conscientiousness, one of the “Big Five” personality traits in psychology.

But there may be a solution in case attention to minutiae might be lacking, and I mean something other than importing Swiss or Japanese artisans. Here is a video of a futuristic, totally automated Japanese CNC machine carving a motorcycle helmet out of a block of pure aluminium, with no human intervention other than by the programmer who coded the software :

[ the obligatory Fez doffing for the video pointer ]

Part 2, England, of my critique of Nick Szabo’s view of industrialisation. This is continued from Part 1, “Chinese workers were cheaper than English horses“.

What do coal, American slave cotton, and the resources of the New World have in common ? All of them were important in the English industrial revolution, but none was crucial to getting it started and none might have been absolutely necessary even in the long run. All of them had second-best substitutes which would have been more costly and therefore lowered output. Their absence would have slowed the pace of industrialisation, and changed England’s pattern of trade and specialisation. But industrialisation would not have been stopped dead in its tracks. And the pace would have been slower only until the coming of other innovations, especially those technologies associated with the “Second Industrial Revolution“.

Horses are not even in the same league as coal and cotton, despite Nick Szabo’s argument in his blogpost, “Transportation, divergence, and the industrial revolution“.

I have absolutely no problem with the conventional idea that transport networks are really really important. I just have a problem with the idea that a technology which diffused in 1200-1300 (the draught horse) with a high degree of substitutability (compared with, say, coal and charcoal) should have such a massive importance in 1800.

Industry Location

Industry location was the most important way of dealing with transport costs. In the earliest phase of English industrialisation (1750-1800), industry moved to the sources of its most expensive inputs. Iron works were located near all three ire ore deposits, collieries and canals/ports. The textile mills were also located basically atop coal deposits and water sources. But until the 1830s coal-powered steam was not widespread in Lancashire’s cotton textile industry anyway, which had relied mostly on water power. See below for clickable thumbnail maps of English industrial concentrations in 1700 and 1800 (source). I wish I had a similar map for 1600, which would show a strong clustering of industry in southeast England.

Transporting coal

Szabo says, “The early industrial revolution was highly dependent on bringing together bulk goods such as coal and iron ore”.

Because the cost per mile of water transport was so much smaller than the costs of land transport, this “last few miles to the mine” problem usually played a dominant role in transportation economics, somewhat analogous to the “last mile” problem in modern cable networks.

The “last few miles to the mine” issue apparently played little role in the real-world economics of English coal in the 18th century. From Clark & Jacks 2007 :

In fact, the transport-cost share of the real price of Newcastle coal delivered to London barely changed until the 1820s !

The major drop in transport costs after 1820 may reflect the earliest use of steamship for coastal shipping. Yet, well before the 1820s, there had been quite dramatic reductions in the cost of overland transport in the rest of the British economy. From Mokyr :

In the particular case of coal — the quintessential high-weight, low-price resource of the Industrial Revolution — there is no evidence for Szabo’s contention that “the early industrial revolution…was very sensitive to small changes in land transportation costs”. Apparently location took care of the issue. So imagine what the cost structure would be like for goods with higher value-to-bulk ratios, like textiles and iron, which were not only located near their sources of power, but also located near ports, rivers and canals.

English transport costs already cheap in the Middle Ages !

Overland transport was already pretty cheap in England by 1300. In “Transport Costs in Medieval England” James Masschaele compares overland transport prices between the 1350s and the 1750s. When adjusted with a consumer price index he finds that costs in ~1350 were 40% lower than in ~1750. But just in case that index is not trustworthy, he converts the costs in each period to wheat prices. The results are not substantially different. This can mean, either that overland transport was really more expensive in 1750 than in 1350, or that wheat was much cheaper in 1750. The latter would imply nonetheless that transport productivity hadn’t improved all that much.

The low cost of mediaeval transport in England is confirmed by other data obtained from a different method : the regional integration of the grain market in England over the same time period.

[Source] In a well-functioning market, the difference in prices of the same good across regions should reflect transaction costs, including costs of transport. The gap was low in England quite early. You do get even more improvement after 1700, which you would expect, given the transport productivity growth outlined above. It’s possible transport productivity growth had a much bigger impact on distribution to consumers rather than the sourcing of inputs.

Transport costs are not magical

“…we reach a dramatic but solid mathematical conclusion: the potential value of a land transportation network is the inverse fourth power of the cost of that transportation. A reduction in transportation costs in a trade network by a factor of two increases the potential value of that network by a factor of sixteen. While a power of exactly 4.0 will usually be too high, due to redundancies, this does show how the cost of transportation can have a radical nonlinear impact on the value of the trade networks it enables.”

You can do a “social savings” calculation for the impact of the improvements to overland transport on the total economy. The growth in output in excess of growth in inputs (total factor productivity) in the British overland transport sector was 1.59% per year between 1700 and 1860. That’s a lot. It amounts to a 12-fold increase during those 160 years. The share of the overland transport sector in British national income was about 6%. TFP growth in the overall economy was about 0.56% per annum over the same period. (It was lower in 1760-1800, higher in 1800-60.) So savings in the transport sector account for about 1/5 of the growth in the total economy in this period. (And, yes, I’m aware of externalities.) [Source for these figures is the latest edition of The Cambridge Economic History of Modern Britain, Volume 1, chapter 13.]

But that’s the economic impact of the improvements to overland transport, not the impact of Szabo’s hobbyhorse. I suppose some enterprising drudge could duplicate for England in 1800 what Robert Fogel did for American railroads in 1890 : assess their impact on the US economy by calculating the “difference between the actual cost of shipping goods in that year and the [counterfactual] cost of shipping exactly the same bundle of goods between exactly the same points without the railroad” but with more canal development. (Fogel 1979) Similar calculations have been done for railroads all over the world (but especially British), steam power, computers, telecommunications, and other “general purpose technologies”. Such a simulation would be much easier for horses.

The effects are always non-trivial but more modest than one would expect from the glamourous repute of the technologies, either because the innovations improve welfare more than add to output (e.g., passengers spend less time on travel ; the iPhone doesn’t necessarily improve my productivity, may worsen it, but improves welfare) ; or because the second-best substitutes are never that bad (contrary to Szabo’s Law of No Second Bests) ; or because the large effects are lagged and smoothed over long periods.

My critique of Nick Szabo’s “horse theory” of the Great Divergence between Western Europe and East Asia. This part is about China in the 18th century. See Part 2 for the general issue of industrialisation and transport costs in England.

Week before last in the comments section at Marginal Revolution, I exchanged words with Nick Szabo about his pet theory about the rise of the West. Then I noticed he posted his last rejoinder at his blog, “Transportation, divergence, and the industrial revolution“. The following is my summary of his views.

On the eve of the Industrial Revolution, the North Sea Basin, especially England, had many more draught animals per capita than East Asia. In particular, the English dray horse powerfully complemented all other forms of transport, acting as a “force multiplier” for movement of goods to market by canal, river, and sea. The combination enabled the transport of goods produced from many more acres of land than other livestock, let alone human portage. Szabo claims 20 : 5 : 1 was the megabeast / oxen / human ratio in “acreage access”.

Industrialisation required bringing together bulk goods such as coal and iron ore from disparate locations, which were prohibitively expensive to move by land. Thus, the “potential for a region to be the first to industrialize, was very sensitive to small changes in land transportation costs”. Szabo then invokes Metcalfe’s Law to assert that even small reductions in the cost of transport in England during the 18th century were multiplied by non-linear-dynamic-exponential processes into…  I’m not sure exactly what. The last time I heard this, Jeff Goldblum was talking about butterfly wings causing hurricanes in the movie Jurassic Park.

By contrast, in China, land haulage depended completely or overwhelmingly on human porters pulling carts or wagons. (Szabo seems to rely primarily on vintage photos and films of rickshaws on the Bund.) Therefore, industrialisation was not possible until the Chinese were freed by technology from dependence on human muscle draught.

Szabo is not saying differences in transport infrastructure — like bad roads or a poorly maintained canal network — made the difference. He’s saying, horses, acting as complements to the networks, made the difference. Even if you were shipping coal 1000km down river or down the coast, you still had land haulage costs from colliery to port and from port to final destination. The land portion may have represented a small fraction of the distance, but those costs exacted an “exponential” impact. (Check out the tremendous emphasis he puts on the horse as a driver of the European-Asian divergence in posts such as this, this, this, and this.)

I think it’s a pretty silly “theory”, but it’s a good excuse to talk about other topics.

First, I can’t find any evidence that in the 18th century that the Chinese and the Japanese were totally or even overwhelmingly dependent on human portage for land transport. No doubt they were much more so than in Europe or the Middle East. But the only actual evidence of animal counts that I could find was a large farm survey conducted by John Lossing Buck, agricultural economist and husband of Pearl S., who reported that in 1930 the number of draught animals per acre in China was approximately half of England’s. But that doesn’t tell us anything about haulage animals during the Qing. (If anyone has any information about it please pass it on.) Nonetheless, carts and wagons drawn by oxen, mules, donkeys and (in the north) horses and camels must have played a part — especially if we are going by vintage photos :

My response to Szabo will make the following points :

1. There’s no evidence, contrary to Szabo’s claim, that the early industrial revolution in England was “dynamically” sensitive to small changes in land transport costs. Industry location was the primary means of economising on them. (See part 2 on England.)
2. Without arbitrarily large magical effects, transport is just another factor of production. If stagnation in transport productivity slowed “Smithian” growth (efficiency gains from more internal trade and more regional specialisation), then marginal cost savings via production technology could have compensated. You can still industrialise with higher transaction costs ; you would just be a little poorer than if you had lower transaction costs.
3. Szabo’s argument is ultimately about energy — horsepower, kilowatts, joules, kilocalories, etc. But he only considers output and not the cost relative to output. The gap between English and Chinese transport costs, even on land, was not nearly as big as implied by Szabo’s supposed 20:1 land “acreage access” ratio.
4. Besides, the English cost structure is not necessarily relevant to the Chinese cost structure. The Chinese would have used more of the input that was less expensive for them.
5. Finally, Qing China’s transport system permitted a well-functioning and -integrated market. It could have supported a nascent industrialisation. At worst, stagnation in transport productivity might have set a ceiling on the scale of modern industrial production in the event that it got off the ground in the first place. But it never did. So transport is a premature, superfluous explanation, anyway.

Then, like now, Chinese labour was cheap

Szabo’s Law of the Second Best says no such thing exists. Thanks to the legerdemain of “nonlinear effects”, all imperfect substitutes for the English megaequid energised on hay and oats would have been fatally inferior. As far as he’s concerned, anything else — pasture-fed horses, oxen, Bactrian camels, mules, donkeys, reindeer, or llamas — would have nipped industrialisation in the bud, no matter how many might have been used to generate 1 horsepower.

But of course there must be a degree of substitutability. Even Langdon, who argued the transition from oxen to horses caused a revolution in mediaeval England, cited evidence that oxen moved more slowly but could pull heavier loads. From Smil here are estimates of the energy supplied by various draught animals :

Insofar as weight is an indication of maintenance costs, mules look pretty good in power/speed/cost terms. (The marginal cost of mules, i.e., the ratio of daily feed-wage to the marginal product of labour in power terms, is pretty low !)

A Chinese worker in the 18th century was much cheaper than an English horse. That must have closed some, if not all, of the horse’s power advantage. I calculate, the unit labour cost (wage relative to productivity) of horse haulage in England, compared with the in extremis case of human-only portage in China, would have been 2:1 in silver-money terms and 3:1 in PPP terms. (See the first post in the comments section for my calculation.) This implies that for any given acre’s worth of goods made accessible by human portage, horse haulage was 2-3 times as cheap/productive. No, this does not take into consideration that using horses probably saved on the number of wagons/carts. And presumably you could not produce 5 hp with 50 men because, as a speculative example, the canal walkways weren’t wide enough. (As with most input substitutes, the isoquant curve for horses/substitutes is convex to the origin, and there is a diminishing marginal rate of technical substitution — but not inordinately.) But the point is, the 20:1 ratio is in no way plausible, especially since the Chinese also must have driven oxen, mules and donkeys.

What about actual costs ? I did a handful of quick calculations for absolute costs of land transport. (Again, see first post in the comments section for the details.) They don’t really mean anything, because these costs should be examined in relation to the total value of the goods transported, but they still tell you how cheap Chinese labour must have been :

• Beijing Road, 1723-35, pole-carrier : 2.3 g silver per ton-mile
• Beijing Road, 1460s : 1.4 g Ag per tm
• Northern Zhili province, 18th c. : 1.52 g Ag per tm
• England 1700-50 : 6.8g Ag per tm

In silver terms, in the first half of the 18th century, the daily wage for skilled Chinese labourers was 2-4 g/day, depending on region ; and, for English labourers, ~11 g in London and ~6-7g outside. (Allen et al.) So at best there were modest differences in cost. Land transport was expensive everywhere before the age of rail and autos.

Szabo has argued that Malthusian pressures did not allow the Chinese to keep many draught animals because land was cultivated as intensively as possible for human consumption, which did not leave much room for nutrient-dense fodder. I guess he didn’t figure that extremely high land productivity in China, which resulted simply in more population, would have made made labour cheap. This is just a roundabout way of saying, capital was relatively inexpensive in England, labour relatively dear. The reverse was true in China.

As I said earlier, the country’s relative cost structure should matter more. Shiue & Keller estimates that the costs of sea versus inland waterway versus overland transport in the mid-to-late 18th century were 1 : 2.7 : 9.5 for China and 1 : 2.4 : 7.1 for England. If the cost of land transport relative to water transport was greater in China than in Britain, then the Chinese would have relied much more on water than the British. Obviously. Which is why internal Chinese trade was concentrated in a chunk of land serviced by rivers and canals, which was about 10 times the size of Britain.

There are also estimates of cost per value of goods transported for river & marine transport. One estimate From Shiue & Keller :

Given all of the above, the most reasonable conclusion is : if the costs of land freight relative to other freight were higher in China than in Britan in the 18th century, then the cause was probably not the source of draught power, but, more likely, bad roads. That’s plentifully attested in the historical record.

In England, improvements to roads and canals during the 18th century largely took place at private-sector initiative.

Qing China had a well-functioning market across large distances

In the 18th century, Qing China possessed a transport capacity to bring a vast amount of goods to market across long distances. At Marginal Revolution, I cited in support a passage from Kenneth Pomeranz’s The Great Divergence, a book whose overall thesis I reject. I repeat it for the benefit of readers here :

Pomeranz’s qualitative assessments are supported by some careful studies. Did China in the 18th century have a well-functioning market ? The best way to judge is to see how similar the prices of a bulk commodity were across regions. Shiue & Keller, “Markets in China and Europe on the Eve of the Industrial Revolution“, examines grain prices in Chinese and European cities across a comparable geographical space.

The above shows the degree of convergence in grain (rice and wheat) prices between cities as a function of distance. The more negative the values on the vertical axis, the less difference there was between cities in terms of their grain prices. (I’m not using the word “correlated” because that is the incorrect term.) The inter-city difference in prices, if they were kept roughly constant by price arbitrage, should reflect the transaction costs of inter-city trade. So you can interpret the gap between the curves as an upper bound on the difference in transport costs between European cities, Yangzi river markets and Chinese provincial capitals.

Chinese market integration compared quite favourably with Northwest Europe as a whole in the 18th century. It’s only at the shortest distance that European integration was stronger, which is consistent with the claim that for the Chinese land haulage was more expensive than other kinds of transport. So the incentive in China was to move goods farther away than in Europe, because of the differences in relative costs of land and water transport between the two.

This study lacked 19th century data for China, but by 1825-49 Northwest Europe moved ahead of where the most advanced region of China had been in 1742-1795. England had also been ahead of everyone by the late 18th century. I should add, a more recent study argues that China’s market integration deteriorated between 1740 and 1820. But it explicitly rules out transport costs as an explanation because the deterioration in markets correlated strongly with provincial borders but not with river networks or postal routes.

Also, Britain, in size and population, was comparable to a middling Chinese province. If a couple of adjacent prefectures were specialised and mutually dependent on trade, then that would be the equivalent of integrating London, Manchester and Birmingham.

The exception that proves the rule : the British East India Company to the rescue of Chinese supply bottlenecks

Some time during the early Qing (1644-1911), farmers in Fujian and Guandong began a more intensive cultivation of rice as well as cash crops like sugar and tea. Cotton was being crowded out :

Marks also has material on the expansion of commerce based on coastal shipping.

So even if China’s own transport capacity had been too limited or too slow to expand, foreign traders and shippers in Canton were present to fill the gap — in spite of the infamous Qing restrictions on foreign trade. Just as now, this could have supported a coastal industrialisation. So transportation cannot possibly have been a check upon it.

Qing China’s transport capacity could have supported a nascent industrialisation. It’s true, all else equal, between any two countries, the one with the lower transaction costs (including transportation) should have more internal trade and regional specialisation, and therefore more gains from trade. But modern economic history suggests, such gains face diminishing returns, and efficiency gains from rapid and sustained technological innovation are much much bigger. That’s true even if there are feedback effects between commercial and technical expansions.

In the 18th century England experienced both kinds of expansion — “Smithian” and “technological”, even if the effect of the latter was much bigger. Who says you must have the same balance between the two as England had ? Almost nobody else did.

Qing China lacked the appropriate technological innovation

In 1800, continental European countries were well behind England but many of them still had had at least isolated episodes of industrialisation. That wasn’t the case with Qing China in 1700-1800. There was cotton textile production in the Yangzi delta (with the raw cotton transported from the north). But it was largely a domestic cottage industry of peasant wives spinning at home as a side activity to supplement their families’ low farm incomes. In this labour-intensive sector, there was no equivalent of modest, early Lancashire inventions as the spinning jenny or the water frame.

According to Joseph Needham, the Chinese “treadle-operated three-spindle spinning-wheel” had been the global technological standard for spinning cotton yarn until James Hargreaves invented the spinning-jenny in the 1760s. But the three-spindle wheel had been invented in the 14th century during the Ming dynasty. Soon thereafter, it was “upgraded” into a 32-spindle water-powered spinning-frame for ramie and hemp. :

But the Chinese did not or could not adapt that for cotton, when it became the dominant and preferred textile. Many people now argue the incentive for mechanisation was absent in China because labour was too cheap. (I reject the ironic “cheap labour” theory of Chinese non-industrialisation, but that must be the subject of another blogpost.)

I also contend, just as English industrial concentration shifted from southeast in the pre-mechanisation era to the Midlands and the north in the early mechanisation era, so Chinese industry might have moved closer to the sources of expensive inputs like coal, horses, pasture, and land suitable for fodder crops, in the provinces to the north of Beijing. Manchuria was still a vast frontier ready for exploitation, and about the size of Pakistan or Turkey. The problem in this case was probably political. The Qing dynasty was Manchurian and would not permit commercial expansion in the barely exploited Manchuria and Mongolia.

Which is why I say Szabo puts the cart before his hobbyhorse. There were much more basic failings in China than the lack of horses.

Also see : Szabo is wrong about England, too.

Just quoting my favourite unintentionally hilarious passage from Said’s Orientalism.

On pages 314-5 of his famous book Orientalism, Edward Said quoted a passage from the Middle East historian Bernard Lewis. He then ‘analysed’ it immediately afterward :

Now, you can criticise Bernard Lewis for many things, but most reasonable people ought to find Said’s textual ‘analysis’ of Lewis’s paragraph… insane. Those who don’t find it insane mark themselves out for what they are — not sensible.

The inability to distinguish the perceived (or just hallucinaed) ‘subtext’ from the prima facie meaning of the text is endemic in certain sections of the humanities, especially those quarters which might be deemed “critical theorists”.

By the way, Orientalism provoked a famous acrimonious debate between Said and Lewis in the pages of The New York Review of Books. Cf. Lewis’s commentary and Said’s and Lewis’s exchange.

Did western industrialisation require American slave cotton ? What coal and sugar might tell us. (Short answer: It’s reasonable and plausible to argue slavery accelerated the industrial revolution, but not enabled it. It’s profoundly unreasonable to say the IR could not have happened without slave cotton.)

In a rather hyperbolical claim, Edward Baptist argued in his book The Half Has Never Been Told that American slave-picked cotton was a prerequisite to the Western world’s escape from the Malthusian trap :

It’s the latest in the long history of attempts to ground the “rise of the West” in slavery, colonies, and/or natural resources. I already criticised the productivity claim above in my earlier blogposts Baptism by Blood Cotton and Plant breeding…drove cotton productivity gains, but I have meant to address the industrialisation issue also. First I have a small detour.

(1)

In The History Manifesto, Jo Guldi and David Armitage claim, in passing, that historians are fond of counterfactual logic, but it is foreign to the “theoretical toolset…that currently dominates corporations and policy”. But most people realise historians (trained in history) almost never engage in counterfactual logic at all. In fact, there’s a pronounced allergy to it.

By contrast, every piece of economic analysis about the past, whether short or long, is implicitly or explicitly counterfactual. What was the effect of the increase in the statutory minimum wage ? What did the Fed’s quantitative easing accomplish ? How big was the impact of railroads on the American economy ? Sometimes there are natural experiments which permit tests, but very often they don’t exist. So each of those questions would be answered by comparing the actual world that unfolded with the hypothetical world where either the event did not take place, or in which a different factor might have been substituted.

Robert Fogel, before doing his famous work on the economics of slavery, had cut his teeth by studying the impact of railroads. He modelled an American economy in 1890 that had not built railroads, but had instead expanded canals and other transport systems. Only by at least trying to answer such questions can you really begin to understand the impact of any particular event in the past. In fact, arguing that just because X happened, it must have been terribly important can be just as speculative as any counterfactual.

(2)

Gregory Clark and David Jacks attempt to assess counterfactuals, in ascending order of speculativeness, about the importance of Britain’s coal deposits to the industrial revolution. First, they consider what might have happened if there had been no improvement in the productivity of coal extraction. (Their answer: not very much difference.) Then they consider what might have happened if Britain had had no coal deposits and had to import them, but from nearby. (The cost of coal would have been higher, but its quantitative effect on the Industrial Revolution would have been limited by some substitution of water and wind power. The lack of coal would have affected some industries more than others, so the productivity gains by industry would have had a slightly different pattern.)

Most speculatively, C&J consider what might have happened if Europe had no coal deposits at all. They do this in order to address the arguments made by Kenneth Pomeranz, but China did have a lot of coal in the Northwest. And even Pomeranz is reduced to arguing that the benefits of making those large deposits more accessible via transport investment were simply not obvious ex ante. So I’ve never found the no-coal-at-all-in-Europe counterfactual particularly necessary. C & J argue nonetheless there was sufficient quantity of Baltic timber to complement wind and water (complete with calculations of the shipping costs of timber…). Energy would have been much more expensive, but British industry would have been more energy-efficient. Britain also might have specialised in something other than the iron industry which was profligate in its consumption of coal. I don’t necessarily endorse this view ; I merely note it. (*)

(3)

After The Economist “withdrew” its review of Baptist’s book they kept the article on display but removed the hundreds of hostile comments the review had provoked. Several commenters acclaimed Eric Williams’s Capitalism and Slavery and condemned the lack of willingness to “come to grips with it”. But that had been done a long time ago !

Thomas & McCloskey, which was collected in Floud & McCloskey, has got a nice overview of the research from the Jurassic age of cliometrics on the importance of the West Indies in the late 18th century. The social rates of return on the British capital invested in the Caribbean sugar colonies were deliberately overestimated, but were still considerably lower than the returns from putting the same amount of money in gilts (British bonds). This was because the prices of imports from the British West Indies were higher for the British consumer than world market prices of the same goods, on account of the preferential access given to BWI producers by mercantilist policies. (Only ginger was cheaper !) Moreover, the British Exchequer doled out considerable sums to protect those islands from rival colonial powers. “…British national income would have been considerably higher ‘if the West Indian colonies had been given away'”.

So all those people who now cannot see past the blood sugar riches in Jane Austen novels are revolted at private fortunes made at the expense of the British public.

(4)

What does all that have to do with the cotton produced by American slaves ? Well, I think any counterfactual about cotton is less speculative than coal or sugar. First and foremost, there were ready substitutes for American cotton, as was proven by the “cotton famine” induced by the American civil war.

After the war broke out, the Confederate States of America embargoed the export of cotton in the belief that Britain and France could be blackmailed into supporting the South. By the time it dawned on them it was a futile strategy, the US blockade would effectively prevent any CSA cotton from reaching Europe. And other parts of the world had lept into action to make up for the American shortfall. Britain’s cotton imports in 1856-70 (Surdam) :

At least from India, there was a fairly quick response to the “cotton famine”, which today would surely be called the “cotton supply shock”. India increased its production quite dramatically, and by 1865 Britain was able to import ~70% of its 1860 level from combined sources. The supply response must have been muted in the early years by the uncertainty of the war’s outcome.

Beckert‘s article (which apparently serves as a precursor to the book soon to be released, Empire of Cotton: A Global History) contains many fascinating details, including the observation that the Russian cultivation of cotton in Central Asia really began as a response to the American famine. Or the earliest impact of the United States on Egypt :

Beckert also makes a big deal out of British cotton merchants encouraging subsistence farmers (not slaves) in Pernambuco in Brazil to grow cotton for the British market. But the data above show the short-term elasticity of supply in Brazil was modest compared with India.

Most of the global response to the American civil war should be interpreted in much the same way as fracking or ethanol becomes cost-effective only when conventional crude petroleum reaches a certain price. The USA, with its high yield, could supply more cotton at cheaper prices than India or Egypt, who could only supply more cotton at higher prices. The countries which sought to replace the lost US cotton were primarily expanding acreage by using new land or shifting out of other crops, not increasing yield per hectare.

When prices fell and American cotton finally rebounded in the 1870s, Indian production receded. The expansion in Brazil and Egypt was more enduring. From Wright :

Of course by the 1880s the rest of Europe was a major consumer of raw cotton, and India began selling more to continental Europe than to Britain. Also, after the turn of the century, Japan industrialised with Indian cotton. But I don’t see any indication that the proportions for India, Brazil, and Egypt in global consumption were very different from the above.

(5)

The reason American cotton had become so dominant in the world by 1850 was both price, which slavery did help lower, and quality, which was thanks to nature. New World cultivars produced naturally longer lint than the Old World varieties, making for superior yarn-spinning productivity ; and southern planters also had spent at least 75 years improving on the yield. During the US Civil War British textile manufacturers repeatedly complained about the staple length of Indian cotton lint and its brittle quality. Clothing made from Indian cotton would look “‘poor and thin’ after washing”, but, worse, it was more expensive to work with Indian cotton :

On the basis of such information the counterfactual of no-American-cotton could be worked out. Indeed there must be more detailed archival information about the costs of working with American versus Indian and Egyptian cotton. It seems like an ideal hellhole for Greg Clark to set one of his students into…

But of course in the absence of American cotton, Lancashire mills would have optimised their machinery to work with Indian and other cotton, which would have mitigated some of the disadvantages noted above. Logan reports plenty of testimony that British mills refused to retool their machines even though they could have worked more efficiently with Indian cotton, precisely because they expected the resumption of American supply and they feared losing their competitive edge against rivals like France.

And the plant breeding experiments would not have sat still. Logan also reports the British East India Company, in the 30 years before 1860, had experimented with the humidity-adapted American cultivars in the more arid Indian cotton-growing areas and had achieved one small success by the 1850s, the “Dharwar American” variety. But then the Indian Mutiny/Rebellion broke out and in its aftermath the British East India Company was no more. Ultimately the improvement of Indian yields and quality depended on the local long-staple cultivar (Broach).

So in the absence of American cotton there would surely have been an earlier, less haphazard, and more organised attempt to exploit India’s native resources. The government of the Bombay Presidency directly involved itself in cotton issues only during the US Civil War. Beckert also reports the wartime hyperactivity of textiles manufacturers’ associations in Britain, France, Germany, and Russia pressurising their governments to do something about cotton sources.

(6)

A more realistic counterfactual would make American cotton available without slaves, under free labour. It is now well established, especially by Ransom & Sutch, that after emancipation ex-slaves were not willing to supply, under the prevailing wage, the same amount of labour as they had been compelled to do under slavery. The decline in labour supply was between 25% to 33%. I’m not sure if higher wages or better tenancy terms might have induced ex-slaves to work more, and, if so, why that did not happen. What ever the case may be, ex-slaves experiencing freedom chose a substantial substitution of leisure for work.

But the point is, even after Emancipation, the American South could still meet at least 1/2 of the world’s demand for cotton. That just lowers the burden on the counterfactual.

So even with all the disadvantages of Indian cotton, and even with the revealed preferences of emancipated American slaves, no reasonable person should imagine that western industrialisation would have ground to a halt in the early 19th century in the absence of slave cotton from the southwest frontier of the United States. The dramatic rise of America’s King Cotton in the half-century before the civil war was not an “absolutely necessary increase” for the end of the Age of Malthus.

(*) Note : Anton Howes adds to Clark & Jacks’s observations in his arguments against Wrigley’s coal hypothesis. I particularly like Howes’s points 7 to 11. The coal hypothesis  relies excessively on the importance of James Watt, but Watt’s inventions were not exogenous. One could make Watt endogenous in an “ordinary” way like Robert Allen. In that model Watt responded to a particular cost structure, viz., coal may have been cheap in England but the existing technology was also wasteful in coal usage, so there was an incentive to reduce its consumption. But even that does not explain, as Howes elaborates in the manner of Joel Mokyr, the prevalence and persistence of innovation across industries and across time in Britain and in Europe. There was a general air of innovativeness starting in the mid-18th century which can’t be explained just by reference to prices and costs alone. Howes’s observations echo some arguments made by Deirdre McCloskey in chapter 22 of Bourgeois Dignity, but I think he has better, more concrete examples. By the way the McCloskey book is full of arguments, chapter by chapter, which I endorse, although I’m not keen on her “bourgeois dignity” thesis itself.

In The History Manifesto, two historians, Jo Guldi of Brown and David Armitage of Harvard, urge their peers to turn away from microhistory and go back to doing Big History in the longue durée tradition of Fernand Braudel. The book also doubles as a rant against the public influence of the natural and social sciences, particularly economics, arguing that historians can do “causal analysis” better. But the contents of the book cast doubt on the authors’ understanding of complex social and scientific issues. 

The modellers’ putsch

Whilst historians slept, Jo Guldi’s ideological enemies took over the world and she doesn’t like it. Historians once spoke truth to power. But starting in the 1960s they retreated from the longue durée into the “Short Past”. As they became absorbed in the inner lives of witches, the testimonies of slaves, and the day-dreams of landless peasants, their role as counsellers to the public on the great questions of contemporary life were usurped by economists, climate scientists, ecologists, evolutionary biologists, psychologists, and other ahistorical peddlers of grand universal models.

These modellers, hopelessly naïve about the institutional bias inherent in their “impoverished array of historical evidence”, concoct thin, static, outmoded, “essentialist”, deterministic, and patently false narratives. Their reductionist mythologies serve the interests of the powerful by monopolising the public debate on inequality, climate change, and world governance, and by stymieing the utopian solutions which history tells them are within human potential.

Indispensable skills of the microhistorians

By contrast, historians are uniquely capable of assessing diverse kinds of information streaming out of the exotic foreign country that is the past. “The arbitration of data…requires talents and training which no other discipline possesses..”, because such “skills…are often overlooked in the training of other kinds of analysts”. “[D]iscerning multiple sources of causality and ranking them” and “examining them from different perspectives and experiences to offer the fullest possible account” of complex phenomena such as climate change or income inequality are the province of historians, not scientists or economists.

Thus the abilities of a Jared Diamond, the interdisciplinarian par excellence with a command of a dozen different disciplines, pale in comparison with the “deep engagement” of a Nathalie Zemon Davis or a Robert Darnton. In “their intense reckoning with archives”, these social-cultural microhistorians of obscure, forgotten lives “had to grapple” with fairy tales, architecture, “old books and their illustrations”, and more conventional kinds of evidence, in order to write about the “shame-inducing charivaris of early modern France” and the “mystifying cat massacres of eighteenth-century Paris”. The “heights of sophistication” thus reached in the Short Past can be put to good use in our public debates about the past and the future.

But the sciences are self-correcting

I’m a big fan of The Return of Martin Guerre and The Great Cat Massacre, but would Robert Darnton or Nathalie Zemon Davis, as cultural historians, really add any more value to public debate than educated lay people in general ?

There have indeed been many thin models of universalising human behaviour based on evidence that’s historically shallow or excessively Eurocentric. But over time these tend to be self-correcting as debates rage within and between disciplines. Natural and social scientists argue about data all the time — about their provenance, their representativeness, their meaning, their limits. And they have always looked to historians with constructive and reliable methodologies for historical information.

Evolutionary psychology, for example, has matured from a field of poorly evidenced rampant speculations to one much more constrained by evidence and internal debate — enough to turn the irascible evolutionary biologist Jerry Coyne away from scepticism to cautious embrace. Steven Pinker took a pretty strong position on the historical trajectory of violence in The Better Angels of Our Nature, but there had long been a “war over war” within and between disciplines.

For decades the simple model of a single migratory radiation out of Africa had been accepted by archaeologists. Initially, genetic data reinforced this model, but the recovery of ancient DNA has shown that all non-Africans have some Neanderthal admixture ; some non-Africans have admixed with “Denisovans” ; and there must have been many hominid species across the Old World that had been undreamt of by archaeology. Thus the provisional paradigms of science keep finding complications and deviations.

Guldi cites the case of Thomas Piketty as powerful confirmation that social science is short-sighted and would be improved by the historian’s perspective. But her own intepretation of Piketty’s place in the intellectual history of economics demonstrates a fairly normal operation of science. An old hypothesis by Kuznets had held that in industrial economies inequality should initially rise with growth but diminish over time. This view was challenged by Piketty using both longer-run evidence and quite conventional economic methodology. (He has been criticised by “heterdox” economists precisely for being so conventionally neoclassical.) Model, empirical challenge, debate, more data brought to bear on the question : isn’t that how things are supposed to work ? But of course Guldi is unaware that the Kuznets hypothesis has been challenged before Piketty. For example, Gallup 2012 surveys the literature on the empirical tests of the Kuznets hypothesis and finds that the initial confirmations based on cross-sectional evidence were challenged as soon as cross-country panel data became available in the 1990s.

Historians can do “multicausal analysis” ? Who knew ?!

Here I gauge the analytical sophistication of Jo Guldi and David Armitage themselves, using a few examples from the book, as a way of suggesting that cultural-social historians of their type are ill-equipped to deal with serious policy issues. Guldi wisely stays clear of trying to argue against evolutionary psychology, except to mutter, perfunctorily, against biological determinism by noting “gender roles and systems of hierarchy show enormous variations in human history”.

(a) Example of shallow causal analysis

Guldi objects far less to climate scientists than to economists. Her problem with the former is mostly that she doesn’t much care for their “species thinking”, because it’s the Western elites and corporations who are to blame for climate change, not “humanity”. Guldi seconds Malm & Horborg‘s view that the Anthropocene only meaningfully began with the industrial era of steam inaugurated by elite capitalists, for which “at no moment did the species vote … either with feet or ballots, or march in mechanical unison”. Furthermore,

This may be a plausible argument for exonerating countries so poor as to produce virtually no carbon emissions, but a bad one for concentrating the blame on a “particular, small subset of western elite families and corporations”. Millions of consumers around the world in the 19th century did, in fact, “vote” for steam by buying cheap cotton clothing or buying other goods transported by rail or ship.

The bottom half of the US income distribution receives <20% of the total income. That’s still more than $3 trillion — a lot of carbon ! Even the bottom 5% of Americans have higher incomes than at least the bottom 60% of the world. This means, the average bottom-ventile American has got a carbon footprint many orders of magnitude greater than any Afghan or Malian villager. The bottom 5% of Germans would consume even more. And, of course, this global inequality of income was in part enabled by the Industrial Revolution.

(b) Example of naive historicism & extrapolationism

Guldi proclaims…

Markandya does have a plot of SO2 emissions per capita versus GDP per capita :

The above was not intended by Markandya as a statistical analysis of the issue, which is why he notes the “steep decline in the levels of SO2 since 1956…, without any serious impact on GDP per capita”.

Markandya did not extrapolate from the much smaller fluctuations in the longer history (unlike Guldi), because he must be aware of an important confound : the falling energy intensity of the economy resulting from technological change or an expanding service sector. During the 19th century, the amount of energy consumed per unit of GDP fell in Western Europe. From Kander, Malanima, & Warde (a book favourably mentioned by Guldi in The History Manifesto) :

Markandya is aware of this :

Nakicenovic (2002) documents the long-term “decarbonisation” (falling carbon per unit of GDP) of the US economy :

Since energy intensity clouds the causal picture, one cannot naively compare the trajectories of emissions and GDP in order to assess the impact of regulations on the latter. One might be on sounder footing in case of dramatic divergences such as after 1956 in the UK, which is what Markandya focused on.

(c) Plain old flakiness

Many passages in The History Manifesto degenerate into outright flakiness. For example, a section of chapter 3 discusses the possibilities of increased international governance. The “power of historical thinking…destabilise[s] conclusions about the best shape of institutions”. And in rapid order we are told, Samuel Huntington and Francis Fukuyama were wrong ; David Graeber has shown capitalism, historically, has used debt to put people into bondage ; the United Nations used to be more important in the 1950s and 1960s ; there has been a recent upsurge in mass protest movements and indigenous peoples’ movements ; etc. And because of these things, we know… well I’m not sure exactly, but it seems alternative, international modes of governance are more feasible than is reckoned by most people today.

Guldi also routinely commits the “genetic fallacy“, the logical error of confusing the origins of a thing with its current use, meaning, or implications. Thus, the staid metric known as “unemployment” was initially

Finally, may I suggest Guldi confuses the neo-Malthusianism of the ecologists, which really applies to all time, and that of the economic historians, which is only about the preindustrial past ?

Persson himself tells you that Clark “admits” human beings restrict their fertility decisions. Since A Farewell to Alms is cited in Guldi’s notes, she might have known Clark has got a whole chapter on fertility arguing that all human societies, contra Malthus himself, exercised “preventive checks” on fertility via delayed marriage, infanticide, birth spacing, contraception, etc.

Clark’s neo-Malthusianism relies heavily on income inequality. When per capita income is higher, this need not translate into increased population because income inequality is also higher. That means, most of the extra income is extracted by the elites and does not necessarily translate into better living standards for the ordinary person.

What’s not to like, for Guldi ?!

( The argument between Persson and Clark was not about the causes of “poverty and want”. The key bone of contention is what constitutes “subsistence income”. Persson interprets it to mean the absolute minimum for survival and Clark, any level of income, high or low, at which birth and death rates are equal. )

In The History Manifesto, historians Jo Guldi, the Hans Rothfels Assistant Professor of History at Brown, and David Armitage, the Lloyd C. Blankfein Professor of History and Chair of the History Department at Harvard, repeatedly misunderstand or misrepresent the research they disparagingly cite in their references — especially the research of economists and economic historians. In fact, the footnotes (technically endnotes) in the book often bear no relation to the authors’ contentions and, in a few cases, state the exact opposite of what G & A claim their sources say. One begins to wonder, have they actually read any of the references they cite ? The critique touches on many issues along the way, including Victorian inequality, the environment & economics, etc.

This post is an adjunct to La longue purée, my general comment on their book. [Edit 13/04/15: I also have a follow-up, Errata dentata, examining the revisions in The History Manifesto.]

Below I catalogue as many as I could find, in a reasonable amount of time, of the errors, peculiar interpretations, and outright misrepresentations of research cited in the references of The History Manifesto. Although the book has got two authors, I will refer only to Guldi as I believe she authored most of the book. [Edit: See note at end]

Guldi criticises many non-historian researchers in this 125-page monograph with 300+ notes, but she reserves a special animus against economists and economic historians. In the narrative she constructs, broad-minded historians work with fuller, richer, multi-faceted evidence to come up with a balanced view of the past, whilst doctrinaire, myth-driven economists form Panglossian interpretations of narrow, reductionist data.

By Guldi’s own description, pages 57-60 of the book, which deal with the divergent views of historians and economists on the subject of inequality and living standards in 19th century Britain, is the showcase exhibit in the contrast between the two groups. But in this section errors, distortions, and misrepresentations of research are particularly dense, and Guldi’s narrative is eviscerated by their exposure. So I start the examination of G&A’s references with this section.

The exact opposite of what the source said !

There are several cases in The History Manifesto where Guldi attributes findings or statements to researchers which are the opposite of what they actually said. One amazing example, cited no less than three times in the book, is Johnson & Nicholas :

Guldi claims these British and Australian (not American) economic historians found the height of prisoners increasing when, in fact, they quite clearly said the exact opposite. From Johnson & Nicholas, the conclusion on page 480 and chart on page 477 :

The paper also cites much previous research on the subject, and any casual researcher could easily find similar observations in the middle-of-the-road source, The Cambridge Economic History of Modern Britain, which starts the section on health and nutrition on page 134 unambiguously :

The exact opposite #2 + hallucinated subtext

The much-maligned Johnson & Nicholas had actually been cited a little earlier in the book, along with confederates :

Nobody cited in note 54 says or implies anything about socialism, higher taxes, or “stagnating social opportunity” in 20th century Britain. This is simply a case of hallucinated subtext.

J & N have been exonerated, but equally disgraceful is Guldi’s citation of The Gifts of Athena. Economic historians of the industrial revolution are divided into wage “optimists” and “pessimists” (see below), and Joel Mokyr is counted amongst the latter. In Gifts, Mokyr argues (pp 125-130) that even if wages did rise for workers, most of the increase may have been compensating for the reduction in well-being due to the transition from cottage industry to factory work. The new work regime implied commuting, loss of leisure, increased regimentation, more unpleasant and dangerous working conditions, etc. Yet Mokyr does not necessarily accept that real wages rose :

And he is supposed to be a prominent representative of a species who fixates on a narrow metric of well-being and pronounces all is peachy !

As for health and nutrition, it would take too long to summarise Chapter 5 of Gifts, but no fair-minded person reading that complicated portrait of changes in public health, nutritional knowledge, household behaviour, and technology, could come away thinking Mokyr displays simplistic optimism. He does mention that despite the general decline in mortality, the infant and child mortality rates in 1900 were about the same as in 1850.

Long 2013 argues social mobility in Victorian Britain with respect to occupations was higher than previously thought, but :

Long finds the above surprising because over the same period Britain, at first a laggard in supplying public education, converged with other western countries. So you can interpret that to mean, either there was as much opportunity in 19th century Britain as there was in 20th century Britain (at least in income terms), or equally little in both.

I don’t understand what problem Guldi has with Long 2005, a dry yeoman-like study of whether rural migrants to the city in Victorian Britain earned more than if they had stayed in the countryside, and if they were more likely to experience greater intergenerational mobility. There is one sentence in which Long mentions the 19th century did not have welfare programmes like the 20th to inhibit geographical mobility. Maybe this one sentence burgeoned like a Triffid in Guldi’s mind and transformed into “twentieth-century ‘socialism’ resulted in higher taxes and stagnating social opportunity” ???

References have nothing to do with claim in text !

Johnson & Nicholas, the gift that keeps on giving ! The other references are equally cited for no reason. Barro 1996, a very famous paper, explores whether democracy induces economic growth. It certainly does not interpret one as a proxy for the other. Madsen et al. 2010 test various theoretical models of the British industrial revolution and conclude that the number of inventors and the fall in population growth are the biggest explanatory factors. Kelly & Ó Gráda 2013 disputes previous estimates of agricultural output and food availability in 1200-1800 and discuss the implications of their own estimates for the nutritional status of the preindustrial English population. [Edit: Kelly & Ó Gráda’s discussion does cover nutritional status in the 19th century as well.] All of them stay very close to their technical tasks and are not particularly romantic about what their results mean. None of the papers cited has anything to do with using wages, prices, or height as proxies for freedom, democracy, or happiness !

Even when the gist is right, a multitude of important errors

Even when Guldi has actually read/understood a paper she cites and has described the gist of it accurately, she still gets important elements quite wrong :

I’ve shown that Guldi’s pp 57-60 are a mine-field of errors and distortions. Now I examine other parts of the book.

Just made up

The idea that Landes conceived his classic book on the Industrial Revolution in “direct support of Green Revolution policies” is substantiated neither in the book itself, nor in the Ashworth reference. Landes is a neo-Weberian who extols the primacy of culture in economic development and, in the particular case of Europe, the primacy of Protestant culture. Ashworth criticises him as part of a neoliberal tradition, including Hayek, which minimises the role of the state and maximises that of the innovative individual in the history of the industrial revolution.

It’s very ironic. Guldi routinely portrays economists as neoliberal freaks, but then when she stumbles on the one historian who happens to be a neoliberal, it goes unnoticed. Maybe she mistook Landes for Norman Borlaug ?

References have nothing to do with claim in text #2

Just when you think it can’t get any worse, Guldi doubles up on the egregiousness :

The papers by Noy 2009 and Noy & Nuaslri 2011 are about the economic impact of natural disasters, like hurricanes and earthquakes, or the fiscal behaviours of governments in the aftermath. They have absolutely nothing to do with technology and growth magically taking care of climate change !!! They don’t even mention climate change ! Storm (2009) summarises a panel discussion all of whose participants agreed that “capitalism’s institutions have to be drastically reformed and made fundamentally more equitable“. Even Goklany, a publication from the libertarian Cato Institute, argues, not that the free market will take care of climate change, but that climate change is not a problem which needs solving, because the net benefits of fossil fuels are far greater.

References have nothing to do with claim in text #3

Guldi cites an economist whom she presents as a revisionist on the environment :

In her discussion of environmental history, Guldi cites historian Paul Warde, but the citation straddles the line between the “opposite of what the source said” and “extremely free and tendentious interpretation” :

Maybe Paul Warde has argued somewhere that the West has long been on a path to “environmental exhaustion” or that early modern Europe had survived an “ecological crisis of unprecedented scale”. And maybe he has elaborated at length somewhere on the “anarchy” in early modern Europe resulting from a “collapsing ecosystem”. But it’s not apparent in any of those sources cited by Guldi. Warde’s “Fear of Wood Shortage“ even ends by concluding “Europe, for all its late eighteenth-century problems, remained distant from any ecological frontier”.

The point in “Fear of Wood Shortage” is that because of high transport costs and political divisions, the market for wood was highly fragmented in Germany and Europe in general before the 19th century, and there could be local scarcities even when there was no general scarcity at all. And there’s the question of whether the rhetoric of “scarcity” was used to justify the political allocation of wood to industrial uses like mining or ironworks. It’s strange that Guldi didn’t pick up on this because, after all, she complains all the time about the enclosures of the commons. And what does Warde describe other than an enclosure of woodland commons ?

Nor does the historiography described by Warde indicate “historians of Germany had documented a crisis in wood”. Rather, he describes a tradition of debate between social-science types (including Marx), who argued that the “wood crisis” was manufactured for political reasons, and the forestry specialists and environmental historians who argued for the reality of such a crisis. In Ecology Warde tells us on pg 228, “historians have tended to rely upon anecdote and second-guessing regulations for at least the first half of the early modern period”, rather than direct documentation of any actual wood shortage. What chapter 4 of Ecology actually documents is that there were many different types of wood for different uses (fuel, building, etc.), and local shortages of some kinds were real enough, especially during the Thirty Years’ War, whilst other kinds were abundant. By any reasonable reckoning, what Warde describes seems less an ecological problem than a problem of political and economic fragmentation in early modern Germany.

Nor, as far as I can make out, does Warde discover “how our ancestors found a way out” through new forms of governance. He seems to have found no evidence (pp 278-79) that woodlands were better managed :

Contrary to Guldi’s assertion Warde even minimises the impact of climate (pg. 96) :

Guldi’s claim that her sources document “energy exhaustion” in Europe by the end of the early modern period is supported only by Malanima’s chapter 4 in Kander, Malamina & Warde. The primary evidence for this is the secular rise in fuel prices in the period 1500-1800, but Malanima acknowledges Robert Allen‘s argument that the real price of fuels did not rise in the period. Only by treating food as energy (which the book does as part of its overall approach) is the argument for “energy exhaustion” tenable. [Edit: Also, since the consumption basket used to deflate fuels prices was composed primarily of food, and food prices were also increasing, the book argues the real price of fuels was actually stable. But it was only stable, because coal was available.]

¿¿¿ Huh ???

I don’t know Jonsson, but I do know the Pomeranz book quite well, and my reaction is total mystification. Class, race, wisdom of local peoples being overridden, rule by landlords the cause of environmental destruction ? None of that is in Pomeranz.

Pomeranz has got a substantial comparison of Chinese versus European land markets and use practises in the 18th century (pp 70-80). And his overall point is that Chinese land markets, on balance, were more efficient than Europe’s. In the latter, ancient entails, hereditary tenancies, and the slow progress of enclosures (e.g., in France, Germany, Spain) compromised the operation of the market and the free alienability of land.

Got that ? If the western elites were complicit in something, it was in retarding the intensive exploitation of land. At least according to Pomeranz. Western Europe on the eve of the industrial revolution theoretically had a lot of slack in land, but customary and institutional barriers kept it underutilised. England was the great exception.

In chapter 5, Pomeranz does advert, on the land that was open to intensive exploitation, to the severity of deforestation (particularly in Britain, France, the Low Countries, and the Mediterranean), plus soil depletion and erosion in most of Western Europe. But again, in his depiction, this seems a result of institutional restrictions on land use. Pomeranz’s bias in The Great Divergence is to argue as much as possible that China was as “Smithian” or “neoliberal” as any place in Europe. So obviously you’re not going to get the perspective of the Movimento dos Trabalhadores Sem Terra. I thought Guldi knew how to “interrogate” sources ?

Outright hallucination

Guldi attributes to Alesina et al. a genetic interpretation of their findings, even though no such thing exists in the paper :

Alesina et al. most assuredly do not argue that “modern gender roles have structured our genes”. (What does that even mean anyway ? The sentence would only make sense if the genes were doing the structuring.) The words “gene” or “genetic” never appear in the paper, and the authors define “cultural norms” as “slowly changing beliefs about appropriate actions in different situations”. So I’m sure the authors would be surprised to hear their paper has been described as supporting a form of genetic determinism. They simply argue that societies which traditionally practised hoe agriculture, as opposed to plough agriculture, do better on metrics of gender equality. The paper also examines the persistence of gender attitudes amongst second-generation immigrants from plough and hoe cultures, and find that the attitudes persist only partially. So the paper, as the authors present it, does not even support the characterisation that anyone is “locked into [their] history” even culturally once they move.

Obviously I could not look at all the notes. But I’m sure there would be more “peculiar” references if one were to look further.

By the way, given all of the above, I find it a minor miracle that Guldi did not think Elinor Ostrom was a strong partisan of the tragedy of the commons, as a opposed to a critic.

I’ve compiled a list of books on economic history (and closely related subjects) at Goodreads. I welcome any suggestions. I also recommend Anton Howes’s Amazon wish list. Both lists were highlighted by Dietz Vollrath at his Growth Economics blog. By the way, I highly recommend his 4-part “sceptic’s guide to [the empirical literature on] institutions“.

Why China did not industrialise before Western Europe may be a tantalising and irresistible subject, but frankly it’s a parlour game. What remains underexplored, however, is the more tractable issue of why Japan managed, but China failed, to initiate an early transition to modern growth and convergence with the West. A recent paper argues that the gap in “state capacity” between Qing China and Tokugawa Japan was responsible for the divergence. [Edit: Please note, this blogpost disputes that argument.]

In the late 19th and early 20th centuries, numerous countries confronted by the military and economic superiority of the European powers launched themselves upon a path of reform and modernisation. The Ottoman empire, well before its abolition by Atatürk, had seen almost a century of fitful reform. Iran’s Reza Shah (père) and Egypt’s Sa’id Pasha were also self-conscious modernisers attempting to remedy their countries’ backwardness and weakness. And there was also the motley collection of events collectively known as the “Late Qing Reform”, whose most famous legacy was the abolition of the imperial examination system. Nonetheless, only Japan managed to drastically alter its institutions and society, import western technology, and generate sustained economic growth.

A working paper, “Asia’s Little Divergence: State Capacity in China and Japan before 1850” (Sng & Moriguchi 2014), proposes some solutions to this mystery. (It was reviewed here, courtesy of Bernardo Batiz-Lazo.)

In this yet another geographical theory of Chinese decline, Sng & Moriguchi argue in effect that China’s size made it intrinsically less governable than Japan. With a bigger territory, the ruler is more dependent on local agents to collect taxes but simultaneously less able to monitor their corruption because of the high costs of transport and information in the premodern era. The ruler must share with his agents the surplus extracted from the peasantry, but cannot increase taxes lest the peasants revolt. More interestingly, they also argue, tax rates will fall even as the economy expands and the ruler will spend less on public goods like infrastructure.

[Sng & Moriguchi construct a dynamic game modelling the interaction amongst ruler, agent, and peasant, in order to suggest this is a general result in geographically bigger polities. Personally I don’t see much value added by this model because the results seem kind of obvious given the assumptions, which include the probability of audit of the agent by the ruler as a decreasing function of geographical size.]

The taxes collected by the Qing state did decline on a per capita basis even as the Chinese economy grew :

In stark contrast, the Tokugawa shogunate was becoming more effective well before the imperial restoration under the Meiji emperor. Because of their greater taxing ability, the Tokugawa could provide more public goods than the Qing in the form of roads, bridges, ports, lighthouses, firefighters, emergency granaries in case of crop failures, etc.

Therefore, according to Sng & Moriguchi, Meiji Japan and Republican China inherited the “governing infrastructures” of their predecessors :

Edit : Note to all people who apparently stop reading here. The preceding was my description of someone else’s analysis. I disagree with the preceding. I argue Japan’s disunity and China’s unity are implicated.

(1)

I don’t find differences in “state capacity”, defined as fiscal capacity, terribly convincing as an important element of the divergence between Chinese and Japanese fortunes in the second half the 19th century. I agree with Bin Wong. Even if late imperial China’s ability to generate revenue and supply services was declining after 1700, the Qing still managed to suppress not only the Taiping rebellion, one of the largest and bloodiest civil insurrections in human history, but also other revolts at roughly the same time (Nien, the Hui, Dungans). Even taking into account participation from the western powers, that speaks to an impressive mobilisation of support and resources despite being in the “decadent” phase.

At the end of 1860, the Qing emperor had fled north of the Great Wall from whence his ancestors had hailed ; the western powers had occupied and partly burnt his capital ; and bizarre pseudo-Christian cultist-revolutionaries had declared a new state in one of China’s most important cities, Nanjing. Yet within a year, the Qing were back. At the very least, the Qing state did not yet face an insurmountable “crisis of legitimacy” in a way they definitely faced after their defeat in war with the Japanese in 1894-95.

Contrast this with the situation in Japan after its forced opening to the West in 1858. The Tokugawa suppressed one rebellion after another, but within a decade the regime collapsed as more and more vassals defected from the shogunal banner. And most of those insurrections issued from the southwestern extremity of Japan. That’s where the rival feudal lords who would eventually overthrow the shoguns and restore the emperor originated.

The Qing comeback after the Taiping is consistent with the very long-run pattern we see in Chinese history. China, stereotypically, experienced “dynastic cycles” with both surging and decadent phases. But these cycles became less volatile over time. The closer we get to the present, “core” China experienced more time under unification than under fragmentation, even during peaks of warfare and violence :

(Source of the graphic: Ma 2011)

Sng & Moriguchi only show taxes up to 1850. But what happened to Qing taxes after the Taiping rebellion ? They rose. An earlier paper (Sng 2014) by one of the co-authors mentions this in passing, but attributes it to the decentralisation and quasi-federalism the Qing were compelled to practise after the Taiping war.

But in reality the Qing simply found other, more indirect sources of revenue. After 1850, they became much more reliant on customs duties as foreign commerce expanded :

[ Source : The Rise of Fiscal States: A Global History 1500-1914 ] Even in real terms, though not in per capita terms, state revenues in the twilight of the Qing were as high as at the peak of their consolidation in the early 18th century.

(2)

The problem with the fiscal capacity approach is it doesn’t explain the right facts about China and Japan. How ever else they may have differed, the most important difference between the late Qing and late Tokugawa when it came to adjusting to modernity, was that a substantial fraction of the elites in the latter was apparently ready, earlier, to embrace radical change in society.

The opponents of the Tokugawa policy of opening to the West were xenophobic and ostensibly conservative. But when the shoguns were actually overthrown and the emperor restored to (nominal) power, the new leaders were anything but conservative or traditionalist. More importantly, a disproportionate share of the Meiji imperial elites came from Japan’s periphery. (More on this below.)

China also had its own kind of imperial reassertion after the Taiping rebellion was crushed — the so-called Tongzhi Restoration. This has been interpreted quite variously over the decades in western historiography. In the 1950s, in the hands of John King Fairbank, the Tongzhi restoration became the standard-bearer of proto-westernisation. But a later generation saw it as the last gasp of neo-Confucian traditionalism. Also, because the Taiping were defeated with vital assistance from local, semi-private militias from Hunan and Anhui organised nominally in the name of the Qing, the reassertion of imperial power has even been seen as a precursor to the warlordism of the 1920s and 1930s. How ever you interpret it, the provincial Han elites eventually rallied to the cause of the Manchurian Qing and the result was quite unlike the Meiji restoration in Japan.

From an economic history point of view, maybe the best gauge of the political attitude between the end of the Taiping war and the Sino-Japanese war is the railway network. The Qing, or factions within the imperial government, would not permit their construction. It was not until after China’s defeat by modernised Japanese armed forces in the war of 1894-95 that the government changed course and tried, haphazardly, a more “Japanese” approach. From Chinese Economic Performance in the Long Run :

[railway map from Wikipedia]

But why this difference in elite attitudes in the mid-19th century ? I have no idea, but it must have something to do with Japan’s longstanding fractious feudalism and China’s tradition of centralised authority. The Tokugawa state even at its zenith was still “weak” in the sense of not having a monopoly on violence or even currency. From Jansen :

The aforementioned domains of Satsuma and Tosa, along with Choshu, overthrew the Tokugawa shogunate in 1868. The ancestors of these vassal clans (daimyo) and other rebel houses enfeoffed in the southwest of Japan had been defeated 268 years earlier in the civil war that put the Tokugawa in power. In other words, many of the very losers in that war remained in their autonomous redoubts and nursed their rivalrous grievances. The Tokugawa controlled them precariously and only with much effort. This is what the political patchwork of Tokugawa Japan looked like :

[Map also from Jansen. You can click to enlarge.] Fudai = “insider” or allies of the Tokugawa coalition which won the Battle of Sekigahara ; and Tozama = “outsiders” or the losers of that battle (or those who remained neutral and declared allegiance to the Tokugawa only after the battle).

(3)

The comparison of China and Japan also goes against the spirit of the “great divergence” scholarship of the last 20 years. Instead of comparing England or the Netherlands with China, historians began comparing single European countries or regions with commensurate “macro regions” in China, such as Lingnan or the Lower Yangzi.

The regional concept matters for “fiscal capacity” as well. Even if imperial China had always been a “low tax regime” compared with Europe, its taxing and spending would not have been uniform throughout the empire. So some of Sng & Moriguchi’s per capita public goods data may be misleading. As a general rule, empires behave differently toward core and periphery. Walter Scheidel, in a comparison of the Roman and Han fiscal regimes, argues they had different patterns of extraction but a more similar pattern of allocation. Rome under both the republic and the principate extracted revenue from its provinces, disproportionately via indirect sources, primarily in order to benefit the Italian heartland. The Han empire both taxed and spent in approximately the same place, the Yangzi valley region. For the Qing, there probably wasn’t much point in building roads in Gansu province (which is bigger than Japan) when most of the taxable peasants whose rice tribute they relied on inhabited the provinces of the Lower Yangzi.

An earlier paper (Sng 2014) by one of the co-authors does argue this very thing for Qing China. It uses the location of the imperial postal network and the prefectural seats to suggest the concentration of state activity. Of course the rulers of China were also keenly aware some regions of China were more prone to rebellion than others.

[Source: Sng 2014]

[Source: Kung & Ma]

(4)

In the final analysis, when you look at the Lower Yangzi macro region, you begin to think, this whole business of a big divergence between Japan and China is exaggerated. From “Economic Growth in the Lower Yangzi Region of China in 1911-37” :

Clearly, the region that had always been the most advanced in China, and the region that is right now amongst the most advanced, was already experiencing “modern” growth comparable with Japan in the interwar period. Japan’s headstart over China in per capita income probably preceded 1800, but the gap with the “Lower Yangzi Macro region” or even the combined Jiangsu-Zheijiang provinces in 1930 would surely have been much smaller, had the attitude of the Qing government or the provincial elites at the end of the 1860s been quite different.  (You could make similar observations at more micro levels, e.g., in sericulture & silk textiles.)

Shanghai itself, as an international treaty port, was administered by a commission of western businessmen, complete with a western legal and institutional framework. How about outside Shanghai ? In the two decades prior to the Japanese invasion, just how much “state capacity” could have existed in the semi-governed environs of Shanghai ? Neither the republican regime in Nanjing nor the various warlords could have had as much “fiscal capacity” as the Japanese. Perhaps it was enough that the region was just a bit more more peaceful than the rest of China, at least before the Japanese invasion.

Edit: I have already heard nitpickings on Twitter that I underplay the Western intervention in the Taiping war (including “Chinese” Gordon and the American Frederick  Townsend Ward) ; and overplay the provincial rallying to the Qing… But the fact remains, the Hunan Army, which took the Taiping capital of Nanjing, was not controlled by Beijing yet was disbanded by its leader. He might have attempted a seizure of power in the capital, yet submitted to the Qing. The Anhui army was not disbanded, but its leader became an important Qing official who would later suppress the Nian rebellion.

The “Hobson-Lenin Thesis”: inequality, ‘underconsumption’, capital exports, imperialism, and the Great War

In a small section in his new book, Branko Milanovic argues that the First World War was ultimately caused by income & wealth inequality within the belligerent countries, resurrecting ideas from John A. Hobson, Rosa Luxemburg, and Lenin. The basic argument: high domestic inequality => ‘underconsumption’ by the masses & ‘surplus’ savings by the elites => capital exports, i.e., search for foreign outlets for investment => the ‘scramble for colonies’ & imperialism => (a major cause of the) WAR.

I examine each element in this chain of logic and reject the “endogenous World War I” view.

First, I do recommend Global Inequality: A New Approach for the Age of Globalization. Its scope is much more cosmopolitan and historical, and more orientated toward poverty and economic development, than other recent books on inequality. But you can turn to the fine reviews by Duncan Green, Martin Wolf, and Francisco Ferreira for information about the book. Or watch Milanovic’s own book presentation, which is also followed by fantastic commentary from Suresh Naidu.

I myself, having a quarrelsome and disputatious nature, prefer to dwell on Milanovic’s ‘endogenisation’ of the Great War that I strongly disagree with. I’ve always wanted a pretext to argue against it. But it gets little attention in either the huge Great War historiography or amongst economic historians, because few people think the idea is valid. Even Piketty made only a passing reference to it. So I’m happy the Hobson-Lenin thesis has been finally exhumed in a well-received mainstream book.

Here is Milanovic endogenising the Great War:

( The argument only takes up ~2 pages in the book, but it’s an important element of Milanovic’s “Kuznets Waves” theory of inequality, which I also don’t think are real. I side with Lindert & Williamson: no general laws of inequality. )

Maybe it’s kind of pointless to debate the causes of a one-off event like the Great War, but you can certainly say what theories do not fit the facts.

And I’ve never thought the Hobson-Lenin imperialism thesis made much sense, on its face. Why do you need “surplus savings” for colonialism and imperialism? The United States of today, which many consider the imperialist par excellence, has been a net capital importer since circa 1980. In the 19th century, Spain, Portugal, Turkey, Russia, the USA (until ~1896), Japan, and Italy were all net capital importers, and all either had empires or had just started major colonial acquisitions. Russia and the USA, as mentioned by Milanovic above, had just brought a big fraction of the world’s acreage under their rule.

IF foreign investment did require colonies and imperialism, weren’t large gross flows enough? You can have zero “surplus savings” but still have huge gross outflows of capital. So I don’t see why there is even in principle a link between inequality and imperialism.

But that’s a minor point. What’s more interesting are the flaws in all the intermediate links between inequality and imperialism in the Hobson-Lenin thesis.

1. Does the ‘maldistribution’ of income explain European capital exports ?
2. Bairoch’s “colonial diktat” is exaggerated.
3. The geography of European capital exports
4. There was no imperialist competition for investment outlets
5. Did imperialism lead to war in Europe ?

Does income ‘maldistribution’ explain European capital exports?

In the half century prior to the First World War, several European countries were large net exporters of capital, but the proper adjective for Britain is massive. Nearly a third of Britain’s net national wealth in 1913 was held outside the country. At its peak, almost half of Britain’s annual savings were invested abroad, implying close to 7% of British national income. By comparison the second and third places, France and Germany, were stragglers.

Hobson attributed these capital exports to the “surplus savings” of the elites and, by implication, the ‘underconsumption’ of the masses. Lenin adopted a similar argument, but with a small difference (which becomes relevant later).

For me it’s not very obvious that inequality => “surplus savings” in the historical context of the 19th century. Yes, all else equal, lower inequality => more consumption, assuming that the lower strata of the income distribution have a higher marginal propensity to consume than the upper strata.

In The General Theory, Keynes criticised Hobson for equating savings with investment without reference to the interest rate which relates the two. [Edit: Thanks to Jo Michell: “Keynes criticises Hobson for for assuming S *causes* I”.] But I think Keynes sort of missed the point, because Hobson was talking about a structural, or long-run, insufficiency of demand which would have been fairly insensitive to the interest rate. And I can believe that in the early 20th century there was a lot of latent demand constrained primarily by income. Nutritional surveys conducted in the UK in the early 20th century suggested “ten percent of the population consumed a diet which was inadequate in all respects, while a further forty percent consumed a diet which was deficient in proteins, vitamins, and minerals; only one person in ten enjoyed a diet which was adequate in all respects” [Floud 2011].

So, yes, in a “Hobson counterfactual” with a large magical income redistribution from savers to consumers, Britain’s capital exports probably would have been reduced. But my Horatian sermon always begins, “Non pares sunt multi ceteri”; there are many ceteris which are not terribly paribus.

First, a portion of the new consumer demand would have been imported. The demand elasticity for grain in 1900 was certainly lower than in 1850, when it was estimated 50% of British workers’ consumption was grain. {?Williamson cite?} But millions of workers with extra income would have eaten not only more grain but also more meat from the New World and probably more continental dairy. And other import goods.

Second, Hobson actually believed British industry ‘overproduced’ and had to sell the ‘surplus’ abroad, and this was the source of the high savings (high profits). But the reality is that by the 1850s (at least) Britain had a deficit every year in the merchandise trade balance. That goes against the stereotype of Victorian Britain as the “workshop of the world”, the titanic manufacturer and exporter who wiped out native industries all across the globe. But the import of food and raw materials consistently exceeded British manufactures exports by quite large amounts.

What made up for it was the surplus in the trade of the famous ‘invisibles’ — the earnings from Britain’s financial, shipping, insuring, and other global services, as well as the profits from its large investments abroad. In fact, these overseas earnings were large enough to finance both the merchandise deficit and the capital needs of the New World.

All this implies that after some point “surplus savings” in Europe were generated not through domestic production but through foreign production i.e., returns from cumulative foreign investments made in the past, not because the domestic economy was producing more than it imported. Simply put, GDP < GNP.

[Edit: I have an elaboration on the above two points in the comment section.]

Third, there were other structural factors which pushed capital abroad, besides inequality — such as demographics and the ‘stage’ of development.

If we go by Piketty’s top incomes data [1, 2, 3, 4, 5], by the early 20th century British inequality was just a little higher than French and German inequality, but the wages of French and German urban unskilled labour were still lower than Britain’s even as late as 1913. Yet France and Germany exported less capital than Britain, with Germany’s foreign investment share of national savings actually trending down (not up) in the 30 years before WW1. More importantly, British and French national savings rates were about the same as American and German, but in domestic investment, UK, France < Germany << USA and the New World in general.

Britain was still more developed than France and Germany, and there may have been diminishing returns to capital. This is where Hobson’s difference with Lenin matters. Hobson thought Britain was “capable of indefinite expansion” if only those “surplus savings” would stay home given enough domestic purchasing power. But for Lenin the rates of return were low because British capitalism was in its death throes.

But if there was a growth slowdown, that was only because, contra Lenin, Britain was at the technological frontier, having already built up its extensive capital stock as a pioneer of the industrial revolution. (McCloskey, as usual, had an amusing way of putting it: Britain could not have had “two Forth Bridges, two Bakerloo lines”.)

Demographics also structurally induced high savings. France, the pioneer of the demographic revolution, was an older country than Britain and Germany, but all three had much lower dependency ratios than the New World — which is where a lot of the capital flowed. Taylor and Williamson (1994) finds that perhaps 2/3 of British capital flows are explained by differences in dependency ratios between Britain and the capital-importing regions.

My point is that inequality alone cannot explain Europe’s “surplus savings”. International differences in dependency ratios and level of development were also crucial structural factors in addition to income distribution. In fact, Hobson may have overestimated the role of the elites and underestimated what he himself called the “thrifty middle classes”.

The imperial trading system: the “colonial diktat” is exaggerated

The “competitive struggle for markets” generated by the elites’ ‘surplus’ wealth, says Milanovic, required creating colonies and introducing the “colonial contract”. The latter is something he has abstracted many times in the past from historian Paul Bairoch’s Victoires et Déboires [e.g. Milanovic (2003)]. At his blog, Milanovic accused Niall Ferguson of misrepresenting the imperial trading regime in The Pity of War:

As part of the same [revisionist] agenda, colonialism has made a comeback, most notably in the works of Niall Ferguson where we learn of “la mission civilisatrice” of  England and France  and such interesting and false factoids that the British colonized area was a free trade zone, while of course India was forced to sell, in what Paul Bairoch aptly termed “the colonial diktat”, all of its raw materials to the metropolis. The latter was thus a monopsonist.  It is indeed easy to pretend that you are a free trader if all that you buy from others is at the discount.

Ironically, Hobson himself would have disagreed with Bairoch as described by Milanovic, since he devoted chapter 2 of Imperialism to arguing how useless were the post-1870 colonial acquisitions for expanding British trade. Hobson even argued that you could let China be carved up by the other European powers, and Britain would still benefit from the Chinese market !

At any rate, it’s completely false that India was “forced to sell…all of its raw materials to the metropolis”. Both British India and Egypt (under the British protectorate) exported raw cotton to Britain’s competitors in textile manufacturing such as Germany, France, and later Japan. Japan started industrialisation partly with Indian cotton and, especially after 1918, exported cotton textiles to British India itself. (I thought all this was now generally known with The Empire of Cotton, which is not exactly a piece of imperialist apologia…)

Milanovic does accurately report what Bairoch called the “ideal type” of the colonial trading regime, but he omits some important details. The economic relationship between the imperial metropolis and its colonies was not inscribed in stone by Moses for 500 years, and the 18th century mercantilist stereotype does not fit very well the unusual period in world economic history that was 1870-1914. This is well illustrated by none other than Bairoch himself, who authored the first chapter of the Mathias-Pollard edition of The Cambridge Economic History of Europe, Vol. VIII, (1989) which contains 25 pages of description of colonial trade policies for 1815-1914.

Here is the quickest summary of the facts, as of 1913, from Bairoch’s chapter:

There was definitely an imperial bias in the trade flows of the colonies (although Mitchener & Weidenmier 2008 argues the “empire effect” was a combination of preferences and lower transaction costs). Another reason for the imperial bias was that the government of the colonies was often the purchaser and importer of inputs for the construction of infrastructure. The Government of India built a national railway system ordering only from British suppliers.

But it is definitely false, pace Milanovic, that “colonies could trade only with the metropolis”. The colonies of the European powers conducted a large fraction of their foreign trade with third countries. Britain in particular did practise — at least after 1849 — free trade with most of its dependent colonies, just as Ferguson said. [“Britain’s policy of free trade meant that there was nothing to prevent German exporters from challenging British firms in imperial markets (and, indeed, in the home market itself)”].

Here is Bairoch’s summary of the colonial trading regimes:

Open door = free trade, or uniform tariff policy for everyone. Assimilated = the colonies were treated as provinces of the metropolis, so the tariff policy for foreign goods was identical for metropolis and colony. Preferential = metropolitan goods were levied zero or nominal tariffs, whereas goods from other countries had higher duties.

This famous diagram is also appropriate:

[Source: Harley & McCloskey in Floud & McCloskey (1981)]

The UK ran large bilateral trade surpluses with India and Australia, which in turn balanced Britain’s deficits with the USA, Canada, and Europe. India settled its deficits with Britain through its surpluses with the United States, Japan, Europe, etc. It’s often asserted, Britain’s maintenance of the classical gold standard relied on its surplus with India; that would not have been possible if India could only trade with Britain !

The geography of European capital exports 1870-1914

The economic history of the 1870-1914 era is tremendously focused on Europe’s epic movement of capital (and people). But it only makes passing references to the Hobson-Lenin thesis, and usually only to say how implausible it is given the geographical pattern of the capital flows.

And it’s true. Colonies may have been acquired and exploited, but they were not the primary destination for the capital exports of the Great Powers. Far from it. The “external investment outlet for the surplus savings” of France and Germany was overwhelmingly Europe, the United States, and Latin America. These regions accounted for ~80% of cumulative French and German investments, with Europe dominant. For the UK, less than 20% of the total went to the so-called ‘dependent colonies’, a category which excludes the rich self-governing ‘white dominions’ which are obviously not part of the Global South today. Most of the 80% of British foreign investment took place in North America, Argentina, and Australasia — in order to financed the capital costs of producing food exported back to Europe !

[Source: Broadberry & O’Rourke (2010)]

As you can see from the above, “finding an external investment outlet for the surplus savings” did not require that the Great Powers be “in physical control of a place”, and actually-existed foreign markets for Western capital were profitable without “other possible competitors … excluded even at the cost of a war”. British India even received French and German investment!

What ever caused this capital outflow out of Europe, it went primarily to help build the railways, the ports, the electricity grids, and other high fixed-cost “social overhead” infrastructure of the fast-growing frontier economies of the 19th century: North America, Australia, Argentina, and Russia.

But the preceding are averages and aggregates over time and may miss trends in capital flowing more toward colonies as imperialist competition heated up before WW1. A good Hobsonist-Leninist might ask, were capital flows toward colonies accelerating?

No.

[Source: Davis & Huttenback (1986)]

Germany’s foreign investment as a share of national saving was falling in the decades before the Great War. French capital exports were rising and the share going to Africa had a modest upward trend, but, in apparent contradiction of the Hobson-Lenin thesis, they were going into the ‘wrong’ colonies — South Africa and British-controlled Egypt ! (More on this below.)

There was little imperialist competition for investment outlets

The lack of “imperialist competition” for investment markets is obvious once you realise how savings actually got distributed from Europe to the rest of the world. Most European capital exports were portfolio investment in sovereign bonds or equity shares in infrastructure projects which were floated in the financial markets of London, Paris, or Berlin. There was relatively little of what we would call “foreign direct investment”. And given the near-complete capital mobility of the 1870-1914, it does not make sense to say there was competition for investment markets. 

• Much of the ‘British’ foreign investment may actually disguise the foreign investment of France, Germany, and other countries, because the British data are based on London capital calls (Davis & Huttenback; Stone). French and German investors often bought securities in London for projects in the British Empire and the Western Hemisphere.

• And most of these foreign capital issues were syndicated, meaning the same bonds or shares usually had their initial public offerings in several markets at the same time. If a British bank underwrote a large bond issuance for a rail project across the pampas in Argentina, normal principles of risk management made it seek secondary underwriters or distributors in other markets.

• These underwriting consortia or collaborations were even more the norm for French and German investors. Even as France and Germany fell into full political and military rivalry following the Franco-Russian entente of 1891-93, there was financial cooperation between their investors, especially in the Balkans and the Ottoman Empire:

“French banks worked in the Balkans with German partners; in Greece, the public loans were subscribed by an Anglo-Franco-German association. In Serbia the B.I.O. and the Berliner Handels-Gesellschaft shared business (such as public loans) in 1895. In Bulgaria an international group under the leadership of the Banque Internationale de Paris (B.I.P.) gave 28 percent of its business to Austro-German elements. In 1894 an understanding between German and French banks divided the percentages of Turkish public loans without bidding. The Deutsche Bank and the B.I.O. agreed in 1905 that 25 percent of the contracts concluded by one would be offered to the other.

“In the great international operations French bankers did not hesitate to practice a true financial collaboration. In many regions of the globe German and French bankers worked hand in hand, respecting scrupulously their zones of reciprocal interest. They took part together in Russian, Balkan, Turkish, Iberian, Scandinavian, and South American business. They grouped themselves in financial syndicates, under the leadership of a Frenchman or a German. The negotiation of large foreign loans provoked numerous contacts among French and German bankers . . . who ignored the frontiers.” [Cameron & Bovykin (1992)]

It is not for nothing that this period in world history is often called the “first globalisation”. There truly was global financial integration, and great power rivalries did not necessarily interfere with it.

Even instances of imperial gunboat diplomacy and “informal empire” on behalf of bond holders which so exercised Hobson — such as when Napoleon III occupied Mexico partly to collect debt in arrears, or when Germany blockaded Venezuela, or when the British navy shelled Peru for the same reason — served the collective interests of the European investor class in general, not the narrow national interest of the country sending the gunboats.

In the case of Russia, which clearly favoured French investors, the cause is obvious and well-documented: the Franco-Russian military alliance starting from the 1890s. Strategic priorities of the state altered the flows of finance; financial needs did not dictate state policy. Finance was forced to serve strategy, not vice-versa.

Nothing better illustrates the virtual irrelevance of the political competition between the Great Powers to their financial collaboration than the behaviour of French investment overseas:

Per Esteves (2011), “African investment [by French investors] is dominated throughout the period by Egypt, but this particular jump [between 1892 and 1900] is two‐thirds explained by the mining boom in South Africa”.

For several decades before the Great War, Britain and France were the chief competitors for colonial possessions, and one of the flashpoints after the British semi-annexation of Egypt in 1882 was over the Nilotic region. Anglo-French tensions were so incendiary they almost went to war during the Fashoda incident (1898). Yet here we have French private investment flowing unceasingly, not to France’s own colonies for the most part, but to the British-controlled parts of Africa. We might even call this the “Fashoda Principle”: politico-military rivalry implied very little about commercial rivalry.

None of the above denies that great power quarrels in the tropics often involved the financial and commercial interests of a narrow class of Europeans. Countries seeking colonial annexations tended to favour territories where their citizens already had some commercial  interests. Some colonial wars were indeed fought, as Hobson argued, on behalf of “sectional interests that usurp control of the national resources and use them for their private gain”. And there were groups in all the belligerant countries who harboured vaguely social-Darwinistic ideas about the need for colonies. But this “crony capitalism” can hardly have been an important cause of the Great War, since “profitable uses for surplus savings outside their own country” were found overwhelmingly not in the tropics, but in the United States, the neo-Britains, and peripheral Europe.

Did imperialism help cause the war?

Almost every book about the Great War lists, virtually pro forma, colonial competition as one of the background causes of the war. Undeniably it created friction amongst the great powers. But I think it’s fair to say that most 0f the Great War historiography also stress colonial competition had very much abated in the decade before the war, unless you count the Balkans.

Ferguson’s The Pity of War has many problems, but one thing he’s very right about is the war that never broke out in the late 19th century between Britain and France, or between Britain and Russia. Anglo-French colonial disputes that lasted for decades had been by far the most heated, and British paranoia about India was continuously incited by Russia’s expansion across Central and East Asia.

Yet, annoyingly, the Great Powers kept on resolving colonial disputes peacefully. The Anglo-French competition eventually came to a halt with the Entente, which was aimed squarely at the European situation. The Anglo-Russian tensions over Persia, Central Asia, the Caucasus, and China were set aside in favour of greater cooperation within Europe itself.

Even quarrels with Germany were largely resolved by the first decade of the 20th century, e.g., Britain dropped its objections to the Berlin-Baghdad Railway, when Germany agreed Britain could control its eastern terminus (again, that Indian paranoia). The anticipated “scramble for China” ended up as the Yangtze Agreement on the open door policy, perhaps the most pristine example of multilateral “free trade imperialism”, with everyone having access to trading and investing in an unpartitioned but prostrate China.

There was — ultimately — just too much European compromise and cooperation in exploiting and carving up the rest of the world.

Furthermore, the “financier parasites” of Hobson and Lenin had simply the wrong interest. Far from instigating European competition in the tropics for self-serving reasons, they feared rivalry amongst the great powers — for the very good and rational reason that they had everything to lose from it. As Britain’s comparative advantage moved farther away from manufacturing to services, the shift also implied that financial capitalism was ever more threatened by great power tensions. British bankers, insurers, and shippers facilitated the prodigious expansion of global trade and investment in 1870-1914 and profited from the economic growth of France, Germany, and the United States. Many UK manufacturers may have feared German trade competition and agitated for protection, much as USA manufacturers reacted to Japanese and later Chinese competition in late 20th century. But the financiers were more powerful. (See this delicious testimony by a Lloyd’s man at the Committee of Imperial Defense.)

The colonial disputes which Britain took most seriously and was willing to go to war over — Egypt (Fashoda), South Africa (German tensions over Transvaal), Afghanistan (Russian relations) — were all related in some way to monopolising maritime access, and eliminating all traces of threat, to India — the one and only ‘dependent’ colony which was economically important to Britain. But all else in the world, including sharing the spoils of China and the Ottoman Empire, was largely open to negotiation.

Except, of course, for the naval rivalry in the North Sea. What actually soured Anglo-German relations was that Germany’s naval programme was perceived as an existential threat to an island trading nation. German dreadnoughts just a “few hours from the English coast” were somewhat more important than Samoa or the Caprivi Strip.

IF the naval rivalry was something crucially important to the outbreak of the Great War, then Avner Offer’s The First World War: An Agrarian Interpretation is much more plausible than the Hobson-Lenin thesis. Both Britain, to a great degree, and Germany, to a lesser but significant degree, were dependent on the import of food, and the North Sea was the choke point for Germany.

( In a provocative review, Milanovic interprets Offer’s book as arguing Britain’s adoption of free trade and its dependence on imported food was the long-run cause of the Great War. The flaw is this: France and Germany protected their agriculture from the flood of North American and Russian wheat, but by the 1880s or so even they became net importers whose deficits were also settled by invisibles, much like Britain. The cause of the overseas food dependency was not free trade per se, as much as relative specialisation in manufacturing, i.e., industrialisation. )

Germany’s motivation would then seem more ‘normal’. After all, the German naval buildup was confined to the North Sea fleet, and it did not seek to match the size of Britain’s navy. So it seems less about Weltmacht than an understandable response for a great power dependent on imports for food and raw materials in a world where war was considered a normal state of affairs. Certainly the naval programme did not imply a luxury comparable with, say, the Soviet navy of the 1970s and 1980s trawling the waters of the South Atlantic or the Indian Ocean.

But IF we are to accept Weltmacht as Germany’s motivation, then the version advanced by Paul Kennedy’s The Rise of the Anglo-German Antagonism 1860-1914 makes more sense. The domestic economy was important, but not in any Hobsonian or Leninist sense (or even in a Fischerian sense). “Most influential Germans believed that their country should secure a large place in the world order” — first and foremost in Europe, secondarily in the rest of the world. Germany’s rulers believed the country’s political standing and national prestige was incommensurate with its sudden and dramatic rise as an economic superpower. In the political affairs of the world, Germany was accounted only a little more than Prussia had been in 1815 — as the junior member of the Concert of Powers. Imagine the chafing if Taiwan, and not the PRC, still represented China on the UN Security Council…

I think that interpretation is most factually consistent with who actually took the decision to go to war in Germany. After all, wars aren’t waged by anonymous ‘deep’ forces. “Structural factors” still require some kind of mechanism exerting pressure on the actual actors. Electoral pressure, domestic interest groups, the colonial lobby, the militarists, whatever.

Hamilton & Herwig (2004), which hones in on the micro-level decision-making process in each belligerant country, describes the civilian jingoistic pressure groups and then concludes:

“There were few direct links between the various pressure groups and the Reich’s small political and military elite. There is no question that men such as [Chancellor] Bethmann Hollweg and [Chief of the General Staff] Moltke were concerned with the views of public opinion-makers. But there is no evidence to suggest that they at any time during the July Crisis shaped their policies according to the perceived interests of those groups. Neither political leaders, nor eminent bankers, nor captains of heavy industry were consulted in the final considerations for war. Nor were Germany’s [jingoistic] academics, theologians, or veterans groups consulted.”

Mark Harrison is blunter:

“No country went to war for commercial advantage. Business interests favoured peace in all countries. Public opinion was considered mainly when the leading actors worried about the legitimacy of actions they had already decided on. If capital and labour had been represented in the Austrian, German, and Russian cabinets, there would have been no war.”

I consider that directly contrary to the Hobson-Lenin thesis: the capitalist bourgeoisie did not have the final power in Germany (let alone Austria or Russia). And the small and specific group of decision-makers is identifiable. One of the arguments made by Fritz Fischer in his second book is that Germany had already taken the decision to go to war in 1912, based on a high-level meeting that year which seemed eerily to reflect much of German behaviour in July 1914. Even if that interpretation is rejected, it’s still telling that only the military and the Kaiser were represented at the meeting, with no civilian government figures like the Chancellor or the Foreign Minister involved.

In all three Germany, Austria, and Russia, a feudal-agrarian-military elite governed over an increasingly bourgeois-industrial society (but especially in Germany). Those decision-makers held the unilateral power to go to war. And they took the decision unaccountably. When it came to matters of war, it’s not even clear that the East-Elbian Prussian Junker class really cared about the opinions of the country’s industrial and banking magnates.

§ § § § §

The Hobson-Lenin theory is weak on the facts. Imperialism hardly required “surplus savings” in the first place. In historical context, income distribution was an ambiguous factor in Europe’s capital flows. Colonial outlets were not necessary for the surplus savings of the European powers. The European Great Powers generally did not compete for investment outlets. The Great War was not supported by financial interests or triggered by imperialist rivalry over colonies.

A follow-up to my previous post, “Did inequality cause WW1? Contra Hobson-Lenin-Milanovic“, elaborating on the inequality=>capital exports link in Branko Milanovic’s overall inequality=>ww1 claim.

In the earlier post, two commenters took issue with my treatment of the inequality=>capital exports linkage. Of course I did acknowledge that high domestic inequality did probably contribute to Europe’s capital exports, in the sense that lower inequality implies higher aggregate consumption and therefore less capital exported. But I also argued that other structural factors (demographics and diminishing returns to capital) should not be underrated.

Yet even IF inequality had been the most important cause of capital exports, then what it helped achieve indirectly was the rapid development of the world’s settler regions and the consolidation of social democracy in Europe.

(1)

Commenter Gaffers suggested that low aggregate demand due to inequality might have lowered the domestic rate of return for British and European investors. That’s an empirical question, but I’m sceptical there could have been so much underutilisation of productive capacity as implied by that suggestion.

In the short-run, domestic and foreign investment tended to fluctuate more or less inversely in the capital-exporting countries. For the UK:

[Source: Edelstein in Floud & Johnson 2004; Top: national savings; middle: domestic investment; bottom: foreign investment]

The swings in the investment cycle were largely driven by differences between domestic and foreign rates of return. Sometimes there would be a big overseas investment boom (in North America, especially), but when overseas rates of return fell, British savings “stayed home”. But what was the balance of the push and pull factors ?

The United States and Latin America offered pretty high rates of return. The gap with the UK (as a long-run average) seems just too large to make ‘domestic underconsumption’ a an important explanation.

(2)

Jo Michell, an economist at UWE Bristol, posted a comment which allows me to clarify something:

The first issue you highlight is that higher consumption [due to a hypothetical income redistribution from British savers to British consumers] would lead to higher imports and lower net exports. But this doesn’t undermine the Hobson thesis: all else equal, lower net exports will also mean diminished requirements for foreign lending….

My observation about extra import demand is connected to the subsequent observation that Britain (as well as, later on, France and Germany) were not trade surplus countries in terms of physical goods. Rather, their current account surpluses were driven by the so-called ‘invisibles’, what today we would call the balance of trade in services and net foreign factor income (NFFI). In other words, Britain, France, and Germany were not producing more (physical goods) than they were consuming at home.

I think this is a problem for any variant of the Hobson-Lenin thesis — if the source of your “excess savings” is actually overseas, and not domestic.

Another issue is the composition of the merchandise trade balance and the pattern of trade which is neglected by purely macroeconomic reasoning about external balances. Britain exported manufactures, and imported food and raw materials. The “Hobson counterfactual” (a large income redistribution from savers to consumers) would have required more food for any year we might be talking about.

But earlier the counterfactual begins, the more the pattern of British specialisation would have been altered. A portion of any extra food demand would have been met by imports, but a portion would have been met through domestic production. Britain’s actual food import requirement was pretty hefty. In 1850, ~25% of calories available per capita were imported, and by 1913 it was <60%. And that’s just the average: lower down the income distribution, caloric consumption would have been lower and the activity level (from unskilled labour) would have been much higher compared to today’s energy requirement.

It’s difficult to speculate off the cuff about the general equilibrium implications of a higher domestic demand for food. But under plausibly reasonable assumptions, that might have:

• slowed Britain’s structural transformation out of lower-productivity agriculture into higher-productivity manufacturing; and
• moderated the improvement in the UK terms of trade (the imports you can buy with your exports). {edit} British TOT underwent an improvement of 30% in the period after 1860, which coincides with the arrival of food from the Americas, etc. {edit}

Both — structural transformation and terms-of-trade improvement — surely had welfare-improving effects in the long run, even holding politics and institutions constant.

I’m not arguing that such secondary effects would have been large enough to offset the effects of direct redistribution of income from savers to consumers in the “Hobson counterfactual”. Surely the direct effect would predominate. But….

(3)

The tendency for domestic production would have been reinforced if the Corn Laws had never been repealed and Britain had continued to protect its agriculture after the 1840s. And I venture to guess, politically, Britain would have stayed protectionist, had it been a more egalitarian country in terms of wealth and asset distribution.

There’s no mystery about why Germany stayed protectionist: despite its powerful industrialists, Germany was politically controlled by an agrarian-military elite of “East Elbian” Prussian Junkers.

Democratic France pandered to the agricultural interest, because the land reforms of the French Revolution had divided up the great aristocratic estates and created a large class of small land holders. In the long run, this may have slowed industrialisation in France, since, even as late as 1945, France was much more agricultural than the UK or Germany.

( Agricultural protection did not seem to slow down the growth of the manufacturing sector in Germany, but it did impede the development of the service sector. As pointed out by Broadberry (1998), US labour productivity converged with Britain’s in the late 19th century not via manufacturing, but via services. In Germany, however, the service sector remained atrophied until after 1945, and its agricultural sector remained large and unproductive relative to the USA and the UK. [Broadberry 2005] )

In plutocratic Britain, the distributive political conflict was between the ascendant northern manufacturers and the traditional landed aristocracy. The aristocracy lost this conflict, and the value of agricultural land slowly eroded over time. But to some extent the agrarian interest was compensated because the landed magnates integrated socially and economically with the global services industry based in the City.

Had Britain been a more democratic country in the 18th and early 19th centuries, and a more egalitarian country in terms of land holdings, the balance of political power might have favoured the agricultural interest. Then we can take much more seriously the idea that greater equality (earlier in the 19th century) might have retarded Britain’s subsequent development.

(4)

Another consideration: by how much would wages not have risen in a counterfactual Britain (and Europe) which did not export the capital to develop the agricultural potential of the settler frontiers? Even as late 1913 French and German unskilled urban labour was paid less than the British equivalent. (See the wage gap for various European countries 1850-1913.)

In the long run, British and European capital exports financed the development of the food exporting sectors of North America, Australasia, Argentina/Uruguay, and Russia-Ukraine. The incredibly cheap food from these sources, as we know from Hoffman et al. (2002), helped lower inequality after 1870 in Europe. Between 1867 and 1910, the cost of living for the bottom 40% of the British population fell nearly 25% on account of lower food prices from agricultural globalisation.

However the effect was considerably more muted in France, which protected its agriculture:

European capital exports, then, should be viewed as an extension of European direct investment in domestic food production. Not having net exports of capital would have slowed down the development of the settler frontiers and checked one of the great equality-promoting forces of the late 19th century.

(5)

And remember, the capital-importing / food-exporting economies of the 19th century were populated by European migrants. British exports of capital and people look almost synchronous in the time series !

[Source: Floud, Johnson, & Humphries (2014); also see the emigration rates per 1000 per decade of other European countries.]

Basically it amounts to this: Northwest Europe’s specialisation in manufacturing entailed a dependence on imported food and raw materials. Demography and industrialisation had produced by mid-century a large supply of young footloose urban workers. In parallel, there were large land-abundant settler-frontiers with a scarcity of labour and capital, high dependency ratios, and a potential for huge agricultural and raw material production. The “first globalisation” connected the two processes — European workers emigrated en masse and European capital chased them en masse. As O’Rourke & Williamson (1999) argue, Europe’s capital exports “during the age of high imperialism can be viewed, at least partially, as an intergenerational transfer” amongst Europeans.

Hatton & Williamson (1998), probably the most comprehensive treatment of the economics of mass migration in 1870-1914, finds that along with demographics (as an independent variable), the most important factor explaining emigration was the gap in real wages between sending and receiving countries. So we might speculate, if there had been less inequality in Europe on the eve of the Great War, not only would there have been less capital export, but also fewer people exported. Yet people exports were important in the long run to the development of Europe’s egalitarian and social democratic institutions.

One simple way of conceptualising the path of income inequality in the course of industrialisation is the Lewis Turning Point, named after the development economist and Nobelist Arthur Lewis. In the early stages of industrialisation, because agriculture is becoming less important as an employer, there is urban-rural migration and the supply of workers rises faster than the urban demand for workers especially in manufacturing.

The Lewis turning point is where migration into the cities decelerates, and employers start bidding up wages, as labour bargaining power is strengthened. (It’s basically the point where the labour supply curve switches from being horizontal to ‘normal’ and upward-sloping.) Before the LTP, wages grow more slowly than labour productivity, and the labour share of income falls; after the LTP, wages grow at the same rate as labour productivity and the labour share rises.

Emigration helped accelerate the coming of the Lewis Turning Point in 1870-1914, especially for Sweden, Norway, Italy, and Ireland, where emigration really thinned the ranks of labour.

How much did emigration, by itself, contribute to wage growth in the people-exporting countries of Europe? In various simulations, Hatton & Williamson conclude

• roughly, wage growth due to emigration for 1870-1910: Denmark 11%, the UK 9%, Italy 33%, Norway 15%,  Sweden 12%;
• for Ireland 1851-1911: “roughly 40 to 50 percent of the observed growth [in non-farm wages] can be attributed to emigration”; and the lower bound for the estimate on farm wages is 25%

Emigration may also have contributed to egalitarian institutions more indirectly. Recent evidence from Sweden suggests that having an “exit option” via emigration emboldened labour activists in the period 1867-1920 and ultimately increased membership in trades unions. Since labour agitation carried the risk of sacking, intimidation, and blacklisting, the option to emigrate reduced the cost of social activism. Scandinavia’s famous social democracy was made possible, at least in part, by emigration — or rather the “first globalisation”.

§ § § § §

The overall point I’m trying to make is that even if inequality drove the export of capital and people, the latter two also contributed to lowering inequality in the long run through the development of the neo-Europes and the relief of demographic pressure in old Europe.

Postscript

There are many echoes of economic history in the recent comeback of ‘underconsumption’ as an idea in the debate about the recession, inequality, and “secular stagnation”.

• Like Keynes but unlike Hobson, Stiglitz argues inequality makes the recovery more sluggish and incomplete.
• Like many historians (now mostly discredited) who blamed capital exports for Britain’s industrial decline, Hobson believed the domestic economy was “capable of indefinite expansion” — which I interpet as a point about long-run growth, not business cycles.
• I pay little attention to contemporary politics but I recently discovered that a minor storm had erupted over the heterodox economist Gerald Friedman. He assumed in his analysis of the economic proposals of the US presidential candidate Bernie Sanders that lowering inequality and increasing aggregate consumption would double the growth rate of productivity in the US economy. Friedman apparently argues not merely that inequality causes GDP to remain below potential, but that inequality lowers potential GDP, something which is not well supported by mainstream economics.
• Friedman appeals to Verdoorn’s Law as justification — something which has been cited in historiographic debates about (a) the supposed harm done in the 19th century to the British economy by exporting so much capital; and (b) the supposed harm done by the erosion of Britain’s share of the world’s export markets after the Great War.
• Finally, there’s a crazy article claiming or implying that the Industrial Revolution was driven by consumption and demand (it was not).

Historian Sven Beckert’s widely acclaimed book, Empire of Cotton, is a good agrarian, business, and labour history of a single commodity. But as economic history it’s not so good.

I think many readers are disarmed by the book’s magisterial sweep across time and space, which obscures or subdues its underlying thesis. Yet when you remove the dense narrative detail, there remains an ambitious polemic about the political economy of global development. So I have tried to capture the essence of that polemic in about a thousand words — without ever mentioning that dread white fiber.

Note: the following are NOT my thoughts. It’s my summary of Beckert’s book.

{Summary begins}

The West got rich by impoverishing the Rest.

“War capitalism” was the violent exploitation of the non-West through piracy, enslavement, theft of natural resources, and the physical seizure of markets. It was not caused by superior technology or organisation. Nor did it rest on “offering superior goodsat good prices”, such as you find in the la-la-land of economics textbooks, but on the “military subjugation of competitors and a coercive European mercantile presence in many regions of the world”.

Europe had little to offer the world other than some itchy fabrics. And oats. So Europeans profited as the global mafioso middleman skimming the freight between Asia, Africa, and the Americas. Europeans shoved aside Asian merchants and controlled Asian producers directly as both political rulers and monopsonist-buyers. The precious goods from the now-impoverished Asian artisans were paid for (at laughable prices) with the only thing of value the Europeans offered, the gold and the silver stolen from the dispossessed natives of the Americas. The Asian goods were then exchanged for millions of African slaves, who themselves would produce in the Americas the only non-bullion stuff that anyone ever wanted from Europeans.

At the same time, increasingly powerful and interventionist European states practised import substitution industrialisation, with protectionist measures shielding domestic infant industries from the stiff competition of superior Asian goods who were subjected to industrial espionage.

“War capitalism” was a precondition for the Industrial Revolution. It created markets abroad. It supplied essential raw materials made by slaves and other bonded labour. It accumulated capital which funded the new industries. And it was the foundation on which were built the institutions public and private which led to the Industrial Revolution. Even the new technologies invented during the IR would have been for nought without the markets first seized and opened by “war capitalism”.

Equally crucial to the Industrial Revolution was a domestic “war capitalism” — a powerful state compelled self-sufficient peasants and cottage industrialists in the countryside into becoming a wage proletariat. Without this there would have been no labour for the capitalists.

Therefore it was not technology but the role of the powerful state which was the distinguishing feature of the Industrial Revolution. New social relations — the transition from personal relations & customary rules to impersonal exchange and anonymous institutions — would have been impossible without this political intervention.

As the industrial revolution progressed, “war capitalism” transitioned to industrial capitalism, at least at home, with its rule of law and formal commercial relations. But “war capitalism” continued abroad. Industrial growth required the power of the imperial state to open markets across the world as the vent for surplus production. The violent opening of foreign markets deindustrialised the Rest by destroying their traditional industries.

The imperial state also crucially aided industrial capitalism by “enclosing the global countryside”. The domestic commodification of labour had to be followed by an international commodification of labour for this process to work. Just as the “market society” had been created in Europe by the powerful hegemonic state, so the powerful imperial state introduced capitalist social relations to the rest of the world.

In the new international division of labour, previously self-sufficient peasants in the Rest were ripped from their traditional modes of living. Risk-averse peasants preferred growing subsistence crops and supplementing their income with handicrafts. They were unwilling to become an agricultural wage proletariat. So the imperial state taxed them; made them tenants and sharecroppers; turned their commons into private plots; built railroads and ports on top of them; and wiped out the market for handicrafts through mechanised competition.

Peasants were compelled to become monocultural suppliers of raw materials to the metropolitan manufacturers. At the same time, they would also become consumers of the manufactured output of the metropolis. The result was dependence on low wages, predation by rural loansharks, and subjection to the vagaries of the international commodities market. Thanks to this coerced integration with global capitalism, millions fell victim to famine in the late 19th century.

None of this global “great transformation” was some ‘natural’ outcome of market forces. It was an invention, a deliberate ‘reimagining’, accomplished by powerful western states.

Western colonial powers then “kicked away the ladder” of economic development by imposing “free trade imperialism” on the colonies and thus foreclosing any possibility of infant industry protection. Import competition did not spur colonial industries to mechanise and compete with imports. Many parts of the non-European world had once had the prerequisites for an industrial revolution, but the missing element was state capacity. It was destroyed by European “war capitalism”.

The colonised peoples could not benefit from the support of an indigenous “developmental state”. When nascent capitalists did finally appear in the colonies thanks to the advantage of low wages, they were actively undermined by the imperial state who catered to the interests of the metropolitan manufacturers. The Rest remained poor and backward.

The emergence of industrial capitalism in the West was therefore fundamentally zero-sum, coming at the expense of its emergence in the Rest.

But this could not remain true forever. Workers in the “Global North” struggled against capitalists to improve their wages and working conditions. It was their political struggle and collective action which led to the improvement in their living standards. This in turn increased labour costs for metropolitan manufacturers, and made the products of the “Global South” more competitive as the wage gap widened further.

But the most important event in the economic resurrection of the Rest was decolonisation. Indigenous capitalists, who had played a crucial role in independence movements, were finally able to get a “developmental state” of their own to look out for their interests. Now the “Global South” could at last practise import substitution and infant industry protection. This precedent had already been powerfully suggested by Japan in the late 19th century, which was never colonised, and whose government could plan and steer the country toward industrialisation, especially by practising “war capitalism” in Asia.

Contrary to many observers, there was a lot of growth and industrialisation in the post-colonial period.

{summary ends}

In other words, the Beckert recipe is:

• start with Marx and “primitive accumulation“, both domestic and international. Marx had gleefully welcomed the imperialists’ destruction of the “semi-barbarian, semi-civilized communities [of Bengal], by blowing up their economical basis, and thus produced the greatest, and to speak the truth, the only social revolution ever heard of in Asia”. This was, after all, an inevitable part of the transition to capitalism and then ultimately to socialism. But Marx was a bit too cheerful about the prospects of capitalism and industrialisation in the Rest…
• so temper that cold-but-optimistic vision with Karl Polanyi‘s more rueful reflexions on the destructive creation of the “market society” in Europe.
• add a lot of Eric Williams, especially as mediated by Joseph Inikori, to bring in the “Atlantic perspective” of African slaves’ contribution to Western development;
• mix a dollop of Friedrich List’s infant industry protectionism, as interpreted by Ha Joon Chang;
• and get the concoction shaken, not stirred, by world-systems-theorists like Immanuel Wallerstein and neo-Marxist dependency theorists like André Gunder Frank.

One might modify Ecclesiastes and observe, “There is never anything new under the sun”.

Since African slavery and the Atlantic economy have gotten so much attention in the reviews of The Empire of Cotton, I have tried to stress the aspects of EOC so obviously inspired by Polanyi, Ha Joon Chang, and various neo-Marxists. But I think the book is really, truly, powerfully pervaded by Chang and Polanyi. Most historians today are neo-Polanyists, since they are too cynical for Marxian optimism and utopianism. And it goes without saying that amongst them the only path to industrialisation is infant industry protection.

EOC does not mention the IMF or structural adjustment, but I suspect Beckert would consider the “neoliberal turn” a form of neocolonialism for developing countries. After the period of rapid industrialisation in the post-colonial period, most countries of the “Global South” were diverted willy-nilly from their “project of industrialisation” by neoliberalism. The only exceptions were India, China, and other East Asia. In fact I am certain Beckert believes the foundations for Chinese and Indian success right now were laid in the post-war period.