Some new stuff

Cohen and Mandler on silent changes to the History Manifesto.

The preliminary program for BHC EBHA is up.

Lindert and Williamson income estimates for colonial America.

Open Access and Predatory Publishing

LSE Impact Blog has an essay by Monica Berger and Jill Cirasella  on Open Access:

“Although predatory publishers predate open access, their recent explosion was expedited by the emergence of fee-charging OA journals. Monica Berger and Jill Cirasella argue that librarians can play an important role in helping researchers to avoid becoming prey. But there remains ambiguity over what makes a publisher predatory. Librarians can help to counteract the misconceptions and alarmism that stymie the acceptance of OA.”

They have some valid points, but there is also much that I disagree with. They spend too much time criticizing Jeffrey Beall for not being sufficiently supportive of OA. In addition, they confuse the issue of low quality and predatory. There are a lot of low quality journals out there, but they do not charge large fees to publish papers on line, they do not advertise that you can have your paper published in a month, they do provide some peer review and editing. They do not face up to the costs of the rush to OA, especially attempts to mandate publication in OA journals.

Open access is not the same thing as predatory. Open access means that people can view a piece of scholarship without having to pay a fee, either directly or indirectly through their school or employer. Predatory journals exist to make money by selling false information. The false information that they sell is that the papers in them have been published in a peer reviewed journal. Academics pay the predatory publisher to say that their paper has been published in a peer reviewed journal; the academics then put the lie into their cvs and their annual activity reports and their tenure and promotion files. After examining a number of these journals I am convinced that it is all too easy tell legitimate publishers from predatory publishers.   The researchers that publish in these fake journals are not being preyed upon; the people that are led to believe that these researchers are publishing in peer reviewed journals are the prey.  Beall’slist is really more of a tool for these people than it is for researchers.

Being open access does not prove that a journal is predatory. Not being open access does not prove that a journal is not predatory. There is, however, a connection between open access and predatory publishers. Legitimate open access journals have created an opportunity for predatory publishers by publishing online and charging fees. Predatory publishers mimic these features, but, unlike traditional journals they have no incentive to provide peer review and editing. Traditional journals have an incentive to engage in careful peer review and editing. They need to get people to buy their journal. The articles have to be good enough that universities, members of an association, or people in the field will be willing to pay to read them. Predatory publishers have no incentive to expend time and resources on peer review and editing. The last thing they want is to have anyone read the articles.  If you read something like this it will only make it harder to tell people that you thought you were publishing in a legitimate journal.

Personally, I do not see publication in traditional journals as incompatible with open access. I noted in a previous post that I went through a recent issue of The American Economic Review and was able to find an open access, or ungated, version of every paper.  In addition, we were hiring this year and pretty much everyone had a website with access to their job market paper. There are often some differences between the “ungated” version of a paper and published version of a paper; if you want to cite a paper you should probably get access to the published version. But if the issue is simply access to research results the ungated version will typically provide this. It seems to me that this general approach existed in economics for a long time. Even before widespread access to the internet economists distributed working papers.  Pretty much anyone who mattered had probably read your paper years before it appeared in print.  There may be reasons why this approach will not work in some disciplines. There may even be reasons it will not continue to work in economics, but advocates for open access journals need to acknowledge the problems they give rise to and the possible alternatives.

More on The History Manifesto

American Historical Review has Cohen and Mandler’s critique of The History Manifesto and Armitage and Guldi’s reply. Cohen and Mandler also have a rejoinder to Armitage and Guldi’s reply. I have previously referred to reviews of the Manifesto by Pseudoerasmus and Mark Koyama.

More on the "new history of capitalism"

The U.S. Intellectual History Blog has published an interesting essay by James Livingston on the new history of capitalism and Walter Johnson's River of Dark Dreams. The essay is part 3 of a 4 part series.

More on the recession of the early 1920s

I saw the other day that James Grant’s The Forgotten Depression had received an award from the Manhattan Institute. The book argues that the economy recovered quickly from a “Depression” in the early 1920s because the government did not intervene, and that this provides a lesson for our times. On the same day I read a new paper in the Journal of American History, “Before the Roar: U.S. Unemployment Relief after World War I and the Long History of a Paternalist Welfare Policy,” by Daniel Amsterdam.  You would think they were written about two different countries. Amsterdam describes numerous government responses (largely at the state and local level, but in some cases promoted at the federal level) to unemployment during the recession in the early 1920s.

I have to say, I find The Forgotten Depression and its reception a bit puzzling. It seems to me that the need to identify examples of times when the economy recovered quickly without government intervention is motivated more by politics than by economic theory or historical evidence. Why should a recovery be quick? If a credit boom leads to a severe misallocation of resources, why would we expect that reallocation after the boom would occur at any particular speed? Where is there in, for example, Austrian Business Cycle Theory a method of predicting how many months a recovery will take? Why, even in the absence of government intervention, might it not take years for reallocation to occur?

I can understand making an argument that, other things equal, markets should adjust more quickly when there are fewer restrictions placed upon them. But 1920 and 2008 are so far from other things equal it is difficult to make useful comparisons. The two periods differ fundamentally in terms of the source of the boom and bust. The misallocation of resources, by peacetime standards, was driven by the war.  In addition, Grant focuses on the federal government’s response to unemployment, but before WWII spending by state and local governments (as well as regulation) exceeded that of the federal government.  Just because the federal government did not do something in the past does not mean that government did not do it. One would need to look carefully at state and local actions to understand the role of “government” during the recession of the early twenties. Amsterdam does not provide a complete picture of state and local action, but it is a good start.

And, yes, I called it a recession, not a depression. If we choose to call the early twenties a depression then almost every downturn in U.S. history should be called a depression. Grant rejects recent estimates of historical business cycles by Christina Romer. Perhaps Romer’s estimates are in error, but I do not find the lyrics of “Aint We Got Fun” to be persuasive evidence that she erred.

The graph below shows percent change in Real GDP (1890-1950) based on the Millennial Edition of Historical Statistics (Ca 9). The recession of the early 1920s was simply was not unusually long or severe compared to other downturns.

 

The History of Corporations and Securities Markets

Leslie Hannah’s keynote address to the Economic and Business History Society, “The Origins, Characteristics and Resilience of the “Anglo- American” Corporate Model,” is now online.

 

Mary O’Sullivan argues that J.P. Morgan’s role needs to be re-examined in “Too Much Ado About Morgan’s Men: The U.S. Securities Market, 1908-1914.”

A Tale of Two Plantations

I have recently written about some new books on the history of slavery that I did not like. I just finished one that I really did like: A Tale of Two Plantations: Slave Life and Labor in Jamaica and Virginia by Stephen Dunn. It is a remarkable work, following the lives of hundreds of slaves on two large plantations: Mesopotamia in Jamaica and Mount Airy on the Northern Neck of Virginia, not far from where I live. The bloggers at the Junto devoted several days to discussion of the book and an interview with the author. One of the things that I most enjoyed was the description of the process of historical research. I thought this book was a great example of the historian taking his reader along with him to the archives, describing the difficulties in finding sources, the limits of those sources, and how he made certain inferences from those sources.  

There is also a website that goes with the book.

Burnard and Baptist on the History of Slavery

I posted a link to Trevor Burnard’s review of The Half Has Never Been Told in Slavery and Abolition. The review raises many of the issues that I and Pseudoerasmus had previously raised in our blogs. Because the journal also published Baptist’s reply I thought I would review my take on the book and Baptist’s reply to some of the criticism.

My understanding of Baptist’s book is that he attempts to develop the following chain of reasoning.

1.      Slaveholders were capitalists, seeking profits by physically coercing, i.e. torturing slaves to produce more.

2.      Over time slaveholders increased productivity in cotton picking.

3.      More intense (or more effective) coercion was the source of increased productivity.

4.      The increases in cotton production were the driving force behind American economic growth.

5.      Therefore, American capitalist development was driven by increasingly intense exploitation of slaves.

 

Link 1. The first link in the chain is the strongest. It is also the least novel. No one disputes that slaveholders physically abused slaves to force them to work harder. The only real disputes have been about the extent to which rewards were used relative to the extent to which punishment was used. Fogel and Engerman leaned toward an emphasis on positive incentives. Research since Time on the Cross, leans the other way. Gutman showed the flaws in TOC analysis of whipping. Rick Steckel demonstrated that heights of slaves were significantly lower than whites, consistent with inadequate nutrition. The available evidence suggests that the amount of labor that African Americans supplied after emancipation was much less than they had supplied under slavery. And, although one might question the extent to which they are representative, numerous slave narratives and autobiographies provide evidence of extensive use of whipping and other forms of torture.   

Baptist, however, also tries to portray his view of slaveholders as capitalists as a shift in historical understanding, and this is one of the points upon which Burnard criticizes his book.

Baptist replies to Burnard

in my introduction, I suggest that historians have not done enough to show the relationship

between nineteenth-century changes in the Southern slave economy on the one hand,

and the emergence of industrial and financial capitalism in the USA on the other. Here

Burnard is miffed that I do not begin with a recitation of historiographical begats and

begottens. True enough, in this trade-press book, I chose to eschew the traditional long

historiographical slog of a monograph’s introductory chapter.

 

But in his book he does not just suggest that historians have not done enough. He declares that

“during the late antebellum years, northern travelers insisted that slave labor was less efficient than free labor, a point of dogma that most historians and economists have accepted.”

This statement is demonstrably false and misleads readers about the historiography of slavery. there have been surveys of economic historians that show that more than two-thirds would agree that slave agriculture was efficient relative to non-slave agriculture. It has been more than a half century since Conrad and Meyer showed that investment in slaves had a return comparable to other potential investments. Fogel and Engerman long ago argued that slave agriculture was as dynamic a version of capitalism as existed anywhere in the United States. In awarding the Nobel Prize to Fogel in 1993, the Nobel committee stated that “Fogel showed that the established opinion that slavery was an ineffective, unprofitable and pre-capitalist organization was incorrect. The institution did not fall to pieces due to its economic weakness but collapsed because of political decisions. He showed that the system, in spite of its inhumanity, had been economically efficient.”  

 

Burnard and others have also questioned Baptist’s use of approaches usually associated with fiction to describe the slave experience. This is not my style and it is generally not what I like to read. On the other hand, as long as it is clear when he is trying to use historical imagination to enhance our understanding I don’t think it is necessarily illegitimate as a technique.

 

Link 2. There is considerable evidence that productivity of cotton production increased. Baptist, however, did not produce this evidence Paul Rhode and Alan Olmstead did. Baptist cites them, but he cites an old working paper rather than the paper that was published in the Journal of Economic History several years ago.

Link 3.  Baptist misrepresents Rhode and Olmstead on the source of productivity increase. They argue that the increase was due to improvements in cotton plants. They not only provide evidence for increased productivity, they provide evidence that increase in productivity were largest in areas where there was the most development of new cotton plants. This evidence is the reason that they argue that plant breeding was the primary source of productivity increase. They did not just assume that it must have been improved knowledge about plant breeding. Baptist uses their evidence on an overall increase in productivity but does not address their evidence that productivity increased at different rates in different parts of the South. 

It is plausible that slaveholders got better at coercion over time. They could have, for instance, made more extensive use of the record books that form Olmstead and Rhode’s primary source to more effectively determine their use of force. The trouble is that Baptist does not show that coercion increased over time or respond to the evidence that differences in plants caused the increases in picking productivity. Link 3 is weak.

 

Link 4.  This link is the weakest. It is also not new. In the 1960s, Doug North essentially argued that cotton exports were the driving force in antebellum growth because they were at the center of interregional trade between the South, the West, and the North. However, evidence since then has accumulated that tended to undermine this theory. The South was not dependent on the West for food. Northern development was driven largely by intra- regional rather than inter-regional trade (see, for instance, Lindstrom on Philadelphia or more broadly, David Meyer on industrialization).

The fundamental problem is that although cotton was a large part of exports, exports were not a large part of GDP. Consequently, cotton only accounted for about 4 percent of GDP. Baptist seeks to deal with this through a bit of imaginative accounting. As I have pointed out previously Baptist simply makes up the numbers for his calculation.

 

He replies to Burnard’s criticism of his calculation

  

Or consider my ‘back-of-the-envelope’ calculation on page 321 of my text. ‘Economic historians

. . . don’t work out GNP by “back of the envelope” calculations’, Burnard huffs. Here is

what two of the finest economic historians of the nineteenth-century USA say about how

they ‘work out’ historical GNP estimates:

All pre-1929 estimates are based on fragmentary data that were not originally collected

for the purpose of making national product estimates. This means that the

series are less precise than the official estimates. Moreover, the further back in

time these estimating methods are pushed, the more degraded the quality of existing

data and the more scarce reliable detailed series become. These problems force the

investigator to fill the gaps with interpolated data, rough estimates, and conjectured

relationships between available and missing data. (PaulW. Rhode and Richard Sutch,

‘Estimates of National Product Before 1929’, in Historical Statistics of the United

States, Millennial Edition On Line, ed. Susan B. Carter et al. (Cambridge: Cambridge

University Press, 2006), 3–12) In addition, careful readers will realize that whatever the strengths or weaknesses of the speculations and conjectures which I undertake on page 321, the exercise is not actually one of ‘working out GNP’.

 

This is a red herring. Interpolating and estimating based upon conjectured relationships are not the same as just making things up. Economists that make this calculations document the methods and the evidence that they use to make these estimates. Baptist just throws numbers out, with no rational or documentation.  In addition, although he says here that he is not working out GNP, in his book he refers to GNP as a measure of economic activity and then concludes by telling us that nearly half of economic activity can be attributed to cotton production by slaves.

 

Without links 3 and 4, the final link in the chain also fails. What we are left with is a book that documents the abuse of slaves and their movement west. These are important and, at points, Baptist handles them well. There isn’t, however, much that is new.

 

As an economist I also feel the need to comment on the exchange between Burnard and Baptist regarding economics. In his book Baptist frequently feels the need to tell people what economists think or say. Almost invariably, he ends up showing how little he knows about economics.  I have already described his trouble with GDP, but here are two additional examples.

“it led to continuous increases in productivity per person- what economists call “efficiency.” Page 112

Productivity refers to output per unit of input. Productivity per person is like saying output per person per person. There are several different definitions of efficiency. None are the same as productivity. In other words, this is not what economists call efficiency.

“For decades before the financial crisis of 2008, most economists dogmatically insisted the behavior of the market and its actors was inevitably rational. Yet a few brave souls insisted that the history of bubbles, booms and crashes showed a clear historical record of mass irrational economic behavior.” Page 270

In most of economists, rationality is a pretty narrow concept having to do with preferences. It requires things like

1.      For any two bundles of goods a person either prefers one to the other or indifferent between the two   

2.      If  bundle A is preferred to bundle B, and bundle B to bundle C, then A is preferred to C

 

There is absolutely nothing in this conception of rationality that implies that people have perfect information or that things will always work out well. Many economic models analyze how things might not work out well if people asymmetric information. A few people may have used the phrase rational market, but it is not the typical use of the word rational among economists. Many economists do refer to efficient markets. Ironically, the efficient market hypothesis implies that markets follow a random walk; they are not predictable.