« Ah. So That's How the Republican Party Works... | Main | Shifting Points of Reference? »

May 12, 2005

Limitations of the Marshallian Toolkit

An (unfortunately still early) draft of my paper for the IFPRI/Cornell conference on threshold effects and nonlinearities in economic growth:

Adding to the Marshallian Toolkit: Big Push and Nonlinearity in History and Theory

J. Bradford DeLong
U.C. Berkeley and NBER
May 2005

ABSTRACT: Most of economics practiced, even today, uses the Marshallian toolkit. But there is an often-undervalued thread of economic theory--with traces in Smith, and visible in writers like Hirschman and Rosenstein-Rodan--that has focused on virtuous circles that, it is claimed, produce powerful non-linearities in growth and development. Looking back at economic history, one cannot be struck by how much of the economic past that is truly interesting is inaccessible to the neo-Marshallian toolkit. My hope, at least, is that a closer engagement with economic history will help economists build alternatives to the Marshallian toolkit, and help specify the situations under which "big pushes" are beneficial in that the whole is greater than the sum of the parts, and under which small changes in incentives may produce powerful changes for good or ill.


This is the first time I have been asked to give a keynote, so I am unpracticed and somewhat unsure how one is supposed to proceed. I suspect that my greatest value-added will come in simply sketching out the importance of the problem that this conference is attempting to address. And the best way to do that, I think, is to point out how very little the smooth continuous-curve economics of Alfred Marshall--he who said that "nature does not make leaps"--can explain of the truly important aspects of economic life worldwide over the past two centuries. We live in a world of big pushes, thresholds, and nonlinearities. Yet we economists try to tackle it with a broadly neo-Marshallian toolkit of smooth changes and equilibrating responses, and so we do not do as well as we should.

There are, of course, theoretical alternatives and additions to the Marshallian toolkit: think of Rosenstein-Rodan (1943, 1961); Hirschman (1958); and Murphy, Shleifer, and Vishny (1989). Krugman (1996) regards Rosenstein-Rodan's (1943) "Big Push story as the essential high development model" in which:

development is a virtuous circle driven by external economies... an interaction between economies of scale at the level of the individual producer and the size of the market.... If modernization can be gotten started on a sufficiently large scale, it will be self-sustaining, but it is possible for an economy to get caught in a trap in which the process never gets going....

But as Krugman says, this "Big Push" branch of theory--even though it can trace itself back to Adam Smith (1776) and "the division of labor depends on the extent of the market--virtually vanished from the face of the earth in the 1960s and 1970s. Krugman attributes this to the fact that:

economies of scale were crucial to high development theory.... [And] economies of scale were very difficult to introduce into the increasingly formal models of mainstream economic theory.... A rise in the standards of rigor and logic [in economics] led to a much improved level of understanding of some things, but also led for a time to an unwillingness to confront those areas the new technical rigor could not yet reach.... After 1960... an attempt to publish a paper like Rosenstein Rodan's would have immediately gotten a grilling: "Why not build a smaller factory...? Oh, you're assuming economies of scale? But that means imperfect competition, and nobody knows how to model that, so this paper doesn't make any sense."... By the 1970s (when I was myself a student of economics)... [Big Push ideas] seem[ed] not so much wrong as meaningless.... Where were the models?...

But the increasing-returns model-building is not that hard. (Indeed, Krugman wonders about his own argument: "[was] the long slump in development theory... really necessary. The model [of Murphy, Shleifer, and Vishny (1989)] is so simple: three pages, two equations, and one diagram... could... have been written as easily in 1955 as in 1989. What would have happened to development economics, even to economics in general, if someone had legitimized the role of increasing returns and circular causation with a neat model 35 years ago? But... economists... were still absorbed in the possibilities of perfect competition and constant returns.... Good ideas were left to gather dust in the economics attic for more than a generation; great minds retreated to the intellectual periphery.) I suspect that the major problems lay elsewhere than in Krugman's diagnosis--which I think was a factor but not the factor. Two more important factors were, first, that investments in the standard neo-Marshallian toolkit promised high returns; and, second, that thresholds, increasing returns, and Big Pushes were scary. Small changes can have big effects, yes, but how then does one know which small changes to study?

As an economic historian, I cannot help with the second problem--although I will have a few words to say about it at the end of the essay. But I can help with the first: I am here to say that the payoffs to investments in the standard neo-Marshallian toolkit are in fact a lot smaller than most economists who do not worry about threshold effects and nonlinearities presume. Indeed, the most interesting parts of economic history are issues on which the neo-Marshallian toolkit has next to no purchase.

This essay documents that claim. In the next section, I briefly discuss how unlikely it is that the neo-Marshallian toolkit could give us much purchase on the biggest problems of development: neither the fact of modern economic growth nor the extraordinary divergence in relative prosperity over the next two centuries yields to the neo-Marshallian toolkit. Section 3 takes a slightly less Olympian view. It argues that three major problems in economic history--understanding the British Industrial Revolution itself, the East Asian Miracle, and western Europe's rapid post-World War II growth--all require serious thought about thresholds. We do not have all the answers, but it is plain that we cannot make progress without incorporating substantial nonlinearities into our models. At its end, section 3 considers a separate, non-historical problem: the macroeconomic collapse of neoliberal reform in Argentina at the start of this decade, and makes a similar argument that the neo-Marshallian toolkit is not of much help.

This paper thus serves a primarily negative purpose: it shows the extent of our ignorance, rather than putting forward strong positive models of how exactly, thresholds were passed and Big Pushes triggered. The concluding section 4 offers an apolog for this, and tries to provide a few insights into how economic inquiry might be able to use models in which thresholds and increasing returns are central in a productive way.

2. The Big Problem

Consider the biggest push of all. We live today in a world of extraordinary and unbelievable wealth. Within two generations human literacy will, if we avoid blowing ourselves up with thermonuclear weapons, be nearly universal. 2300 years ago Aristotle said that a society of even narrow literacy--let alone a society in which philosophy in its oldest sense of the love and pursuit of relatively abstract knowledge--was impossible in the absence of widespread slavery, or something like it, and that that would never change unless we went back to the Golden Age in which there were looms that could throw the shuttle by themselves and plows that could dig the furrows by themselves. We don't quite have those--although I did last year hear a presentation by a Caterpillar executive about prospects for teleoperated bulldozers that could be controlled from places of air-conditioned comfort. But we are close enough. What involuntary servitude exists in the world today is by and large an exercise of power over those who have none by those who have little, and not (as it was in the Classical Athens of Aristotle, the Imperial Rome of Hadrian, or the Federalist Virginia of Jefferson) a key distributive institution in the economy as a whole.

We have acquired this extraordinary per-capita wealth in the past three centuries. Before three centuries ago the overwhelming effect of technological progress--and there was a great deal of technological progress, from the mechanical clock and the watermill to the cannon and the caravel to the development of strains of rice that you can crop three times a year in Guangzhou and the breeding of merino sheep that can flourish in the hills of Spain--was to increase the numbers of humanity, not to raise median standards of living. Today, if we divided up what we produce worldwide equally, would it give us a standard of living ten times that of the bulk of our preindustrial ancestors? Twenty times? A hundred times? Does the question even have meaning? David Landes (1999) likes to tell the story of Nathan Meyer Rothschild, richest man in the world in the first half of the nineteenth century, dead in his fifties of an infected abscess. If you gave him the choice of the life he led as the finance-prince of Europe or a life today low-down in the income distribution, checking-in items as they arrive at the loading dock at the local Target, with thirty extra years of life to see his great-grandchildren, which would he choose?

We also live today in an extraordinarily unequal world. There are families today near Xian, in what was the heartland of the Tang Dynasty Empire, with two-acre dry wheat farms and a single goat. There are families today with four-bedroom two-story houses in Tacoma Park and civil-service jobs that could buy the two-acre wheat farm and the goat with one day's wages. Yet in terms of communication, transport, and shipment, the world today is far smaller than the world of three centuries ago when a passage from Britain to India could take half a year, and the London headquarters of the East India Company's directives to its agents on the spot were always at least a year out of date.

Marshall's economics--the equilibrium economics of comparative statics, of shifts in supply and demand curves, and of accomodating equilibrating responses--is of next to no help to us in accounting for any of this. Why, worldwide, did median standards of living stagnate for so long? Why did their rate of growth undergo an acceleration that is extraodinarily rapid in such a short period of time? Where is the economics of invention, innovation, adaptation, and diffusion anyway? Not in Marshall. And why is today's world so unequal--so unequal that it is hard to find any measures of global distribution that do not show divergence at least up until the 1980s (see DeLong (1986))?

It was Robert Solow (1957) and Moses Abramovitz (1956) who most strikingly pointed out that Marshall's economics is of little help in understanding modern economic growth. The action comes not from smooth supplies-and-demands and the allocation of scarce resources to alternative uses, but from the technological-and-organizational-change residual--and that is, after all, nothing but a measure of our ignorance. Mankiw, Romer, and Weil (1989) made a valiant attempt to shore-up the Solow growth theory version of Marshall, and attribute relative wealth and poverty in the world distribution of income to virtue (having a high investment rate and a low population growth rate) and vice (having a low investment rate and a high population growth rate) respectively. But Hsieh and Klenow (2002) undermine this by asking: are countries relatively poor because they have a low rate of investment (and thus little capital), or do they have a low rate of investment because they are poor (and so face a high real price of capital goods)? And that countries are still in the high-population-growth middle of the demographic transition is also both cause and conseuqence of relative poverty. The best that can be said is that the Mankiw, Romer, and Weil (1989) story is the top half of a vicious-virtuous circle--one assigned priority because the authors conduct their investigation under the maintained hypothesis that investment rates and population growth rates are determined exogenously.

Truly exogenous variables to identify cross-country growth patterns are very hard to find. But it is hard to think about the determinants of population growth and relative price structures without concluding that there is potential for the circles of reciprocal causation to be very strong indeed, with infinitesimal changes in outside variables having large and measurable effects.

3. The Problems of Detail

Things get little better when we examine the past several centuries from a less Olympian height, and try to look at instances where it at least looks as though there has been a highly successful Big Push. Once again the things that we find of greatest interest are the places where the neo-Marshallian toolkit helps us the least. Consider, briefly, three positives and one negative: the British Industrial Revolution, Western Europe's extraordinary post-World War II growth spurt, the rise of East Asia, and--the negative one--the crash of Argentina's neoliberal reform program.

The Industrial Revolution

Consider, first, the Industrial Revolution itself. This great threshold nonlinearity is, in many ways, the greatest of them all. Economic historians like Ken Pomeranz (2000) point out truthfully that back before the Industrial Revolution differences in median standards of living across the high civilizations of Eurasia were relatively small. A peasant in the lower Yangtze Valley under Kangxi in the late seventeenth century had a different style of life than his or her contemporary peasant in the Thames Valley under Charles II Stuart, but not one that was clearly better or worse. Two centuries later that was no longer the case: by the end of the nineteenth century median standards of living in Britain and in other countries to which the Industrial Revolution had spread were, for the first time in recorded history, light-years above any neo-Malthusian benchmark of bio-sociological subsistence. Britain's early industrial-era economic accomplishment took place in spite of the drawing-off of a substantial proportion of national income to support a corrupt, decadent, and luxurious aristocracy (see Namier (1929)); in spite of a tripling of population that put extraordinary Malthusian pressure on the natural resource base underpinning the economy (see Wrigley and Schofield (1993)); and in spite of the mobilization of a prevously unheard-of proportion to national income for a near-century of intensive war against a power, France, with three times Britain's population (see Brewer (1989), Williamson (1984), Temin and Voth (2005)).

But the explanation of this extraordinary nonlinearity remains contested. Rostow's (1958) non-Communist manifesto boldly speculated that all that was needed was a market economy coupled with the doubling of society's investment share. North and Weingast speculated that all that was needed was a government strong enough to enforce private property rights and yet limited enough to respect them--and have been criticized by others pointing out that the eighteenth and nineteenth century British state's extraordinary powers of eminent domain and its eagerness to exercise them did play a key role and are definitely not part of the picture of the night-watchman state. Kremer (1991) argued from the principal that "two heads are better than one" that human brainpower was bound to someday attain its critical mass, boost the rate of productivity growth, and elevate standards of living enough to trigger the demographic transition and so cause a phase transition of the economy from neo-Malthusian medieval and early modern dynamics to the dynamics of modern economic growth. But at the key moments of the Industrial Revolution English was a language written and read fluently by at most a million people, and Chinese was written and read fluently by at least ten million. If two heads are that much better than one, shouldn't ten heads be even much, much better? Debate over even the quantitative shape continues. There are those in the ascendant (Crafts (1985), Harley (1982)) who see the Industrial Revolution as narrowly-focused on key transformative leading-sector industries: coal, cotton, machinery, and steam. Such an Industrial Revolution would be ripe for explanation in terms of the narrow logic of technological creativity in the context of the abundance of particular resources like coal, cotton, and iron ore. Others in eclipse--perhaps temporarily (see Temin (1997))--see the Industrial Revolution as an extremely broadly-based phenomenon, and so ripe for explanation as the result of political, social structural, or cultural factors.

We can build models of such a transition as the Industrial Revolution. But these models must have threshold effects. Nonlinearities. The rate of growth of British technological capabilities does not rise smoothly as world population (or even British population) increases. The level of productivity does not inch upward at each generation at a pace only a little bit faster than it had done before. Even the Crafts (1985) and Harley (1982) interpretations that see a "little" Industrial Revolution see it as made up of earth-shaking changes in the key leading sectors--they just see those key leading sectors as, initially, a relatively small part of the entire economy. And, of course, there is the most famous contemporary assessment of the power of the Industrial Revolution, even though it was written from the middle Rhine Valley, a place definitely on its fringes. In its authors' view, the Industrial Revolution:

has shown what man's activity can bring about. It has accomplished wonders far surpassing Egyptian pyramids, Roman aqueducts, and Gothic cathedrals.... The bourgeoisie, during its rule of scarce one hundred years, has created more massive and more colossal productive forces than have all preceding generations together....

Why? How? The neo-Marshallian toolkit is of little help here.

Post World-War II Europe in the Argentine Mirror

Consider, second, a more cheerful story: the story of Western Europe's post-World War II economic and political miracle. Before World War II, the countries of continental Western Europe appeared to be lagging further and further behind the United States. Brutally bitter labor relations, sharp social cleavages, intense distributional struggles that threatened and in many cases extinguished democracy--the interwar political economy of Western Europe as narrated by historians like Maier () sounds a lot like the post-World War II political economy of countries like Chile, and Argentina.

The aftermath of World War II threatened to make things much worse in Western Europe. Its export earnings were gone. Required repair and rebuilding needs were enormous. Inflows of private capital were nonexistent. In the absence of large-scale American aid--by some mechanism like the Marshall Plan--Western Europe faced a future in which hard-currency imports would be meted out by an eyedropper, reconstruction would be prolonged and slow, and distributional fights would once again destabilize its politics.

It is true that in a well-functioning market economy, it is hard to argue that resources for repair and rebuilding would remain critically scarce for long. The European economy had substitution possibilities, and market economies are very good at finding and using them where they exist.

Eichengreen and Delong (1991) take a back-of-the-envelope look at the most severe post-World War II Western European bottleneck: coal:

In 1938 Western Europe consumed 460 million tons.... It produced only 400 million tons in 1948.... [During the] Marshall Plan, Western Europe imported... seven percent of its coal consumption from the United States. Assuming... half of national product... produced in coal-burning sectors... [with] fixed [production] coefficients... elimination of coal imports would have reduced Western European total product over the duration of the Marshall Plan by no more than three per cent.

But would the market economy have been allowed to find the substitution possibilities? The memory of the Great Depression was still very fresh, and predisposed politicians toward intervention and regulation. Governments might well have severely constrained market mechanisms by continuing wartime controls (indeed, the British government tried to do so). The late 1940s and early 1950s might have seen the building-up of bureaucracies to allocate foreign exchange, and to impose price controls to protect the living standards of urban working classes. In short, post-World War II Western Europe might well have been a lot more like post-World War II Argentina.

Post-World War II Argentina did adopt demand stimulation and income redistribution coupled with a distrust of foreign trade and capital and a bias toward the use of controls as allocative mechanisms rather than prices. Diaz-Alejandro (1970) is still the classic analysis of Argentina's post-World War II long-run economic stagnation. As Eichengreen and DeLong summarize his interpretation:

[T]he collapse of world trade in the Great Depression [had been] a disaster of the first magnitude for an Argentina tightly integrated into the world division of labor.... Argentina continued to service its foreign debt... [but Britain and America] took unilateral steps to shut it out of [its traditional export] markets. The experience of the Depression justifiably undermined the nation's commitment to free trade. It was in this environment [that] Juan Perón gained mass political support. Taxes were increased, agricultural marketing boards created, unions supported, urban real wages boosted, international trade regulated. Perón sought to generate rapid growth and to twist terms of trade against rural agriculture and redistribute wealth to urban workers who did not receive their fair share....

This Perónist program was not prima facie unreasonable given the memory of the Great Depression, and it produced almost half a decade of very rapid growth. Then exports fell sharply... as the consequences of the enforced reduction in real prices of rural exportables made themselves felt. Agricultural production fell... low prices offered by government marketing agencies. Domestic consumption rose. The rural sector found itself short of fertilizer and tractors. Squeezed between declining production and rising domestic consumption...

Thus Juan Perón's government had only unattractive options: (i) devalue to bring imports and exports back into balance and meanwhile borrow from abroad (reducing the living standards of his domestic political base and abandoning a great deal of his appeal to national price); (ii) contract the economy, raising unemployment and reducing consumption, and expand incentives to produce for export by decontrolling agricultural prices (but this is even worse, and there is not yet a functioning IMF to blame for the policy turnarounds); (iii) control and ration imports to keep domestic demand growing. Not surprisingly, Perón and his advisors chose the third option, hoping that a dash for growth and a reduction in dependence on the world economy was good for Argentina. But the long-run consequences were dire indeed. As Díaz Alejandro writes:

First priority was given to raw materials and intermediate goods imports needed to maintain existing capacity in operation. Machinery and equipment for new capacity could neither be imported nor produced domestically. A sharp decrease in the rate of real capital formation in new machinery and equipment followed. Hostility toward foreign capital, which could have provided a way out of this difficulty, aggravated the crisis...

Subsequent governments could not fully reverse these policies, for even after his overthrow the political forces that Perón had mobilized still had to be appeased. The 1950s saw capital goods become hugely expensive in Argentina--and thus even a healthy savings effort produced little in the way of real investment. To quote Diaz-Alejandro again: "the capital... in electricity and communications increased by a larger percentage during the depression years 1929-39 than... 1945-55." Unable to invest on a sufficient scale, the Argentine economy stagnated.

In 1929 Argentina had been richer than any large country in continental Europe. It was still as rich in 1950, when Western Europe had for the most part reattained pre-World War II levels of national product. Eichengreen and DeLong conclude that slow Argentinean growth thereafter came because Argentina's "mixed economy was poorly oriented: the government allocated goods, especially imports, among alternative uses; the controlled market redistributed income. Thus neither the private nor the public sector was used to its comparative advantage."

Why didn't Western Europe follow a similar post-World War II trajectory? Díaz Alejandro argued that four factors set the stage for Argentina's astonishing post-World War II relative decline:

  1. a politically-active and militant urban industrial working class,
  2. economic nationalism,
  3. sharp divisions between traditional elites and poorer strata, and a government used to exercising control over goods allocation that viewed the price system as a tool for redistributing wealth rather than for determining the pattern of economic activity.

From the perspective of 1947, Western Europe looked as vulnerable as Argentina to economic stagnation induced by populist overregulation. The war had given Europe more experience than Argentina with economic planning and rationing. Militant urban working classes calling for wealth redistribution voted in such numbers as to make Communists plausibly part of a permanent ruling political coalition in France and Italy. Economic nationalism had been nurtured by a decade and a half of Depression, autarky and war. European political parties had been sharply divided along class lines for two generations.

Yet Europe avoided this trap because of small differences that mattered. Large-scale American aid to boost the pace of reconstruction and create, earlier, the impression that the liberalized mixed-economy was working? Fear of Stalin's T-34 tanks just east of the Fulda Gap and a belief by nearly all--conservative and social democrats alike--that this was not a good time to sharpen distributional conflicts? From the standpoint of 1945 it is not easy to see what would lead one then to forecast the wide gaps between Argentina and company on the one hand and France and Germany and company that we see today.

Once again, small differences that appear to matter a lot.

The East Asian Miracle

Last, consider East Asia. It is hard for us today to cast our minds back to the 1960s, and to realize how broadly unexpected the rapid economic rise of East Asia over the past generation and a half has been. From today's perspectives, studies like World Bank (1993) or Rodrik (1994) look back at East Asia and see a host of favorable initial post-World War II initial conditions, from favorable demography and low income inequality to high education levels and relative ethnolinguistic integration as playing a large role. Even Rodrik (1994), who wants to conclude that skillful government interventions (i.e., providing subsidies to producers that win the export tournament) can do significantly better than laissez-faire feels a need to caution his readers: South Korea and Taiwn shared a number os special initial conditions... high levels of educational attainment... equal distribution... that... other countries lack.... [T]he relevance of their experience with government intervention to other developing economies may well be limited. The World Bank's (1993) East Asian Miracle shades the story more toward the initial conditions side. And expressions of official orthodoxy like the Asian Development Bank's 1997 Emerging Asia: Changes and Challenges can flatly assert that East Asia is rich today because of "a fortunate combination of initial potential."

This would come as a surprise to those on the ground in the 1950s and 1960s examining that potential. Berkeley undergraduate thesis student Charlene Huang (2005) cites William Easterly (1995) reporting a World Bank mission's conclusion at the start of the 1960s that there is "no doubt that the development program [proposed by the government] far exceeds the potential of the Korean economy.... It is inconceivable that exports will rise as much as [forecast]..." And, of course, South Korean exports and GDP growth exceeded their targets.

Indeed, Huang (2005) claims that regressions suggesting a powerful role for initial conditions in East Asian development characterize it as:

having high initial levels of human capital (Rodrik 1994).... Primary school enrollment rates circa 1960 are...employed.... However, a country’s primary school enrollment rate circa 1960 is less of an "initial [post-WWII] condition" than an indicator of the competence and success of [governance]... a reflection of the... government's commitment to education and human capital development than as an indicator of initial levels of human capital.... Barro and Lee’s data on the percentage of the adults over 25 years of age in 1960 who completed primary school is... a [better] indicator.... The percentage of adults who completed primary school during the time of colonization is a better indicator of colonial legacy.... A comparison of primary enrollment ratios in 1960 and the Barro-Lee data on the stock of human capital in 1960 clearly illustrates the difference... Singapore and Korea both have about 100% primary enrollment in 1960, [but] the percentage of the adults over 25 years of age in 1960 who completed primary school was 26.2% for Korea, but only 5.6% for Singapore.... Singapore’s British colonizers were not as interested in educating the masses...

The East Asian tradition of heavy investment in human capital is, for many of the high-performing Asian economies, quite recent. That their governments' abilities to create and manage impressive educational systems has played a key role is not in dispute. But the ability to successfully make a Big Push for education from a near-standing start is a social capability that was, in many high-performing Asian economies, created after decolonization--rather than being a long-standing historical legacy.

Many other countries in the 1960s tried to build high-performing mass school systems, and failed. East Asia succeeded. What alternatives to the neo-Marshallian toolkit will help us understand how?

Modern Argentina in the European Mirror

Consider, last, the recent history of Argentina. Of all the disappointments of neoliberal reform in the 1990s, perhaps the most disappointing was the catastrophic 2001 financial crisis in Argentina. Coming after four years of slow recession, it pushed output per capita down to a full quarter below its late 1990s peak and inflicted extraordinary economic pain on a country that only a few years before had been the toast of card-carrying neoliberals like myself, and the example that we pointed to to show the benefits of neoliberal reform programs.

What the Argentinean government did wrong is well known (see Mussa (2002), Blustein (2005)). It failed to reform its tax collection system to reduce evasion. It failed to either put the provinces on their own bottoms as far as financing their expenditures was concerned, or to gain effective control over provincial-level spending. It failed to recognize that the hard external peg of the peso's value provided by the currency board required that confidence be maintained that the debt-to-GDP ratio would be stabilized. Without confidence in a stable debt-to-GDP ratio, a hard currency peg was an incredible and an unsustainable policy--and there is nothing worse than an attempt to make a credible commitment to an incredible and an unsustainable monetary policy.

In the end, when the immovable object of the currency board met the irresistible forces of a government unwilling to balance its budget and currency speculators--both domestic and foreign--sure that they had a sure-win one-way bet, it was the currency board that crumbled, and carried Argentina into a recession twice as deep relative to the size of the Argentinean economy as the Great Depression was in Argentina.

But what surprised me then and surprises me now is the speed of the collapse of the currency board. When it collapsed, Argentina's consolidated debt-to-GDP ratio was about 50%. That is not an unsustainable debt load. And the Argentinean government was managing to run a primary surplus. If there had been confidence in Argentina's fiscal future--confidence that no financial crisis was on the horizon--then interest rates would have been much lower, and the primary surplus would have generated only a moderate general deficit. With low interest rates, Argentina's prospects for growth would have been relatively good. With good growth prospects and a relatively moderate overall government budget deficit, there would be no reason to fear that fiscal policy is unsustainable. Only the fact that a crisis was expected pushed interest rates up to the level where investment was strangled, growth impossible, the overall budget deficit large, and a crisis inevitable.

Michael Mussa (2002) sharply criticizes the fiscal-policy dynamic analyses carried out by the IMF staff a year and more before the collapse:

What should be made of this analysis? The general principle in such matters is that if you are prepared to swallow the assumptions than you should also eat the conclusions. Or, as we used to say when I was growing up, "If, if, if... If my grandmother had wheels, she would be a bus...

This is, I think, too harsh. For the alternative, good equilibrium was out there somewhere: the equilibrium in which domestic and foreign speculators have confidence, interest rates stay low, policies appear successful, the government can leverage that apparent success into larger primary surpluses which further increase confidence, and has a decade or more to get the political-economy ducks in a row to ensure long-run fiscal balance and so medium-run rapid growth. We know that that good equilibrium was out there: Lula da Silva just to the north in Brazil, in a country with equally intractable long-run problems of macroeconomic management and much worse problems of income distribution and public management, appears (cross your fingers) to have found it. But the Argentineans could not.

Here, too, the neo-Marshallian toolkit is of little use.

4. Conclusion: What Is To Be Done?

Suppose that we accept the belief that Alfred Marshall and his toolkit no longer help us where we need help. Perhaps it never did. Perhaps it is just that the mother lode of problems accessible to this toolkit is now nearly mined out. In either case, what do we do now? This would be a far better paper if it could do more than point out that the truly important and interesting parts of economic history are filled with Big Pushes, thresholds, and nonlinearities. But unfortunately all I can do is offer scattered observations.

First, Harry Johnson (1971) observed in his not-quite-fair critique of Milton Friedman's monetarism that a successful intellectual revolution in economics had to do two things: (i) provide the young with a reason not to have to spend their time studying the work and doctrines of the old, and (ii) provide the young with something interesting and straightforward to do. In this, Johnson wrote, both Keynes and Friedman succeeeded admirably. The General Theory claimed that there was simply no point in reading pre-Keynesian business cycle theory. Friedman's monetarism claimed that the only relevant macroeconomic equation was the money-demand function (and perhaps some price adjustment process): all else was illusion and deception. Young Keynesians could go estimate consumption, investment, and export functions. Young monetarists could go estimate money demand functions and think about the division of nominal output into prices and quantities. Alternatives simply fell by the wayside. The endogenous business cycle tradition of Joseph Schumpeter and Wesley Mitchell required that the young spend too much time learning the doctrines of their elders, and so delayed the beginning of their own research. The American institutionalist tradition did not provide the young with something straightforward and productive to do: thus John Kenneth Galbraith is the last American institutionalist.

A successful economics of thresholds and nonlinearities will have to meet these challenges, especially that of providing the young with clearly productive things to do. On the theoretical side, there needs to be a grammar, or perhaps a hierarchy, of threshold models worth investigating. On the empirical side, there need to be consensus views about which potential ways of modelling are in fact likely to be fruitful.

Second, the stakes are quite large--and always were very large. The neo-Marshallian toolkit really is not sufficient. The chances of long-run success are greatly raised by keeping economists' minds focused on all the truly important problems where the neo-Marshallian toolkit makes little progress.

Third, success is not guaranteed: just because economics needs models of virtuous and vicious circles to help it understand the world doesn't mean that journal editors will publish articles using them, or that graduate students will become excited by them. The new toolkit must be shown to be of great use even before it is built, which is hard to do.


Moses Abramovitz (1956), "Resource and Output Trends in the US since 1870," American Economic Review.

T.S. Ashton (1968), The Industrial Revolution 1760-1830 (Oxford: Oxford University Press).

Paul Blustein (2005), And the Money Kept Rolling in--and Out: Wall Street, the IMF, and the Bankrupting of Argentina (New York: Public Affairs).

John Brewer (1989), The Sinews of Power: War, Money, and the English State, 1688-1783 (New York: Knopf).

Gregory Clark (2001), "Debt, Deficits, and Crowding Out: England 1727-1840," European Review of Economic History

N.F.R. Crafts (1985), British Economic Growth during the Industrial Revolution (Oxford: Oxford University Press).

J. Bradford DeLong (1986), "Productivity Growth, Convergence, and Welfare: Comment," American Economic Review

Carlos Diaz-Alejandro (1970) Essays in the Economic History of the Argentine Republic (New Haven: Yale University Press).

Barry J. Eichengreen and J. Bradford DeLong (1991), "The Marshall Plan: History's Greatest Structural Adjustment Program"

C. Knick Harley (1982), "British Economic Growth Before 1841: Evidence of Slower Growth," Journal of Economic History 42:2, pp. 267-89.

Albert Hirschman (1958), The Strategy of Economic Development (New Haven: Yale University Press).

Albert Hirschman (1967), Development Projects Observed (Washington: Brookings Institution).

Albert Hirschman (1970), Exit, Voice, and Loyalty (Cambridge: Harvard University Press).

Chang-Tai Hsieh and Pete Klenow (2002), "Relative Prices and Relative Prosperity"

Charlene Huang (2005), "Can Favorable Initial Conditions Explain Rapid Growth in East Asia?" (Berkeley: B.A. Thesis Draft).

Harry Johnson (1971), "The Keynesian Revolution and the Monetarist Counter-Revolution," American Economic Review

Michael Kremer (1991), "Population Growth and Technical Change: One Million BC to 1990," Quarterly Journal of Economics.

Paul Krugman (1996), "The Fall and Rise of Development Economics" http://web.mit.edu/krugman/www/dishpan.html

David Landes (1999), The Wealth and Poverty of Nations

Charles Maier (1988), In Search of Stability (Cambridge: Cambridge University Press).

N. Gregory Mankiw, David Romer, and David Weil (1989), "A Contribution to the Empirics of Economic Growth," Quarterly Journal of Economics.

Alfred Marshall (), Principles of Economics

Karl Marx and Friedrich Engels (1848), Manifesto of the Communist Party.

Kevin Murphy, Andrei Shleifer, and Robert Vishny (1989), "Industrialization and the Big Push," Journal of Political Economy 97: pp. 1003-1026.

Michael Mussa (2002), Argentina and the Fund: From Triumph to Tragedy (Washington: IIE).

Louis Namier (1929), The Structure of Politics at the Accession of George III.

Ken Pomeranz (2000), The Great Divergence (Princeton: Princeton University Press).

Dani Rodrik (1994), "Getting Interventions Right: How South Korea and Taiwan Grew Rich" (Cambridge: NBER).

Paul N. Rosenstein-Rodan (1943), "Problems of Industrialization of Eastern and South-Eastern Europe," Economic Journal.

Paul N. Rosenstein-Rodan (1961), "Notes on the Theory of the Big Push," in Economic Development for Latin America (London: Macmillan).

Walt Whitman Rostow (1958), The Stages of Economic Growth (Cambridge: Cambridge University Press).

Adam Smith (1776), An Inquiry into the Nature and Causes of the Wealth of Nations (London: Millar).

Robert Solow (1957), "Technical Change and the Aggregate Production Function," Review of Economics and Statistics 39: pp. 312--320.

Peter Temin (1997), "Two Views of the Industrial Revolution," Journal of Economic History

Peter Temin and Hans-Joachim Voth (2005), "Credit Rationing and Crowding Out during the Industrial Revolution: Evidence from Hoare's Bank, 1702-1862" (Cambridge: MIT).

Abbott Payson Usher (1929), A History of Mechanical Inventions (Cambridge: Harvard University Press).

Jeffrey Williamson (1984), "Why was British Growth so Slow During the Industrial Revolution?" Journal of Economic History 44 (1984), 687-712.

World Bank (1994), The East Asian Miracle (Washington: World Bank)(.

Wrigley and Schofield (1993), The Population History of England ().

Posted by DeLong at May 12, 2005 11:49 AM