Prosperity and Depression

The study came out in Spring 1937 as the League of Nations report on Prosperity and Depression with a “Systematic Analysis of the Theories of the Business Cycle” (part I), based on a survey for which Haberler had corresponded with more than sixty experts, and a “Synthetic Exposition Relating to the Nature and Causes of Business Cycles” (part II).

In his survey Haberler cites Neisser in various passages ([1937] 1964, pp. 53, 62n1, 65n1, 85–86n2) as a critic of the overinvestment theories of Ludwig von Mises and Friedrich Hayek and of the neo-Marxist theory of imperialism and its underconsumptionist arguments, yet classifies him as an advocate of a “monetary under-consumption theory” (p. 134).

In part I Haberler sets the focus on overinvestment theories (chapter 3), ^{Footnote 2} and it is here that we find the first consideration of international aspects of the business cycle. The section on the monetary overinvestment theories (section A, represented by Mises, Hayek, and other “Neo-Wicksellians”), concludes with § 8 on “International Complications.” Haberler describes these as modifications of cyclical movements through the balances of payments, conditioned by exchange rate regimes, exchange controls, capital flows, and the composition of exports and imports. His description gives the impression that it is difficult to develop any definite theory of international causes and effects of business cycles. Yet, Haberler ends section A, § 8, by asserting that the “almost endless multiformity of the international complications” can be understood “as special cases which can be brought under the general doctrine” (p. 71).

What is “the general doctrine”? Absent other indications, one is bound to take chapter 12, on “International Aspects of Business Cycles,” as an attempt to provide a general perspective. Closing the “Synthetic Exposition” (part II), it is preceded by chapters on the definition and measurement of cycles, on expansion and contraction, and on the turning points. While the style of argumentation in those chapters is clearly synthetic, with many references to the underlying literature, the chapter on international aspects appears to be more of Haberler’s own making. The contemporary works on the subject are all listed in the first footnote, starting with Neisser (Reference Neisser1936), ^{Footnote 3} and there are few other references in the remainder of the chapter. The international aspects of business cycles are framed in a structure reminiscent of the treatise on international trade that Haberler had just completed before he started to work on Prosperity and Depression.

Haberler’s Theory of International Trade (1936) is well known for its reformulation of comparative cost theory in terms of opportunity cost. In the introduction and in chapter VII (on “International Trade and General Equilibrium”), Haberler demands that the theory of international trade “be regarded as a particular application of general economic theory” (1936, p. 8). Marginal analysis and the “laws of supply and demand” need to be applied to “the specific assumptions which characterize international trade”: namely, (i), “that different money circulates in different countries,” giving rise to “the problem of foreign exchange”; and (ii), that labor and capital are not mobile across national borders (p. 9). This approach is akin to the modern practice of introducing frictions and imperfections into general equilibrium analysis in order to explain cyclical fluctuations of economic activity. Cycles are understood as products of deviations from the Walrasian reference case of perfectly competitive markets.

In chapter 12 of Prosperity and Depression, Haberler chooses a similar procedure. “In setting out the international complications [of business cycles] in orderly fashion” ([1937] 1964, p. 407), he presents two methods of exposition. The first is a bottom-up procedure, starting from a number of isolated economies that subsequently develop links with each other by trade, capital flows, etc. These links can then be investigated as channels of transmission of cyclical impulses across the borders—“spillovers” and “feedbacks” in modern terminology. The alternative procedure is “to start with the hypothesis of a spaceless closed economy embracing the whole world and to introduce one by one the circumstances which divide and disintegrate that economy.” The disintegrating factors are divided in “natural and inevitable physical facts” and “artificial human devices,” so as “to exhibit the operation of the earlier factors without the later factors but not the later factors without the earlier” (ibid.).

In Haberler’s classification, transportation cost is the basic factor that leads to imperfect mobility of goods and services and, hence, to international complications of business cycles. The second factor is imperfect mobility of capital, and the third is national currency autonomy. To see how these complications modify Haberler’s synthetic theory of the business cycle, a brief sketch of the latter is in order. ^{Footnote 4}

Haberler’s Synthetic Theory of the Cycle

In Haberler’s view, “a general theory” that is capable of explaining the typical sequence of prosperity (expansion) and depression (contraction), with crisis and revival as the turning points, needs to be based on “two regular features of the cycle” (p. 275). One is the “parallelism of production and monetary demand”: i.e., procyclical movements of the price level. The other is constituted by “specially wide fluctuations in producers’ goods”: i.e., an accelerator effect of a change in the production of consumer goods on investment (pp. 277–282). The parallel movements of production and demand are described in terms of Wicksellian loanable funds theory, in which the money supply is endogenous to credit demand, determined by the profit expectations of the borrowing firms and their lenders in the financial sector (pp. 289–339). “[T]he acceleration principle,” according to which “investment is explained by a previous change in the demand for the product” (pp. 250, 304–311), determines the cumulative character of expansion and contraction.

In his synthetic propositions about the turning points that end the upswings and downswings, Haberler emphasizes the interaction of “organic counterforces” and “accidental disturbances” (pp. 346–347). A crisis can thus be caused by endogenous maladjustments in the growth process that reduce the elasticities of the supply of capital goods, bank credit, and base money. These inelasticities create inflation, thereby making the system increasingly vulnerable to deflationary shocks (pp. 355–377). Revivals, in turn, are explained by the emergence of idle capacities in the course of the depression. These increase the elasticities in the supplies of goods and loans, making the system more sensitive to stimulating influences, such as public works and other policies aimed at reducing unemployment (pp. 377–405).

In this context, Haberler introduces a specific international aspect. Analyzing the scope for stabilization policies, he draws a dividing line between small open economies and economies that are sufficiently big to have “a free hand” in their monetary policies. Small economies tend to come under greater pressure to cut wages, forcing them to devalue their currencies in order to mitigate the contraction. Big economies can more easily eliminate the contractive effects of wage cuts by combining “a policy of wage reduction with expansionary measures such as public works financed by inflationary methods. The effect will be to forestall any decrease in total wage disbursements, and consequently in the demand for consumers’ goods, which might otherwise result from the wage reduction” (p. 405).

International Modifications

How do transport costs, imperfect capital mobility, and currency autonomy fit into Haberler’s synthesis? Conceptualized as disintegrating factors, they might be expected to reduce spillovers and feedbacks of cyclical movements among different parts of the world economy. In Haberler’s terminology, they would have a “localizing” rather than a globalizing effect. However, the results of his analysis are not that simple. He shows that, in certain constellations, all three factors spread and exacerbate the cyclical fluctuations rather than limiting them.

The uneven geographical distribution of resources—“whether by nature or, in the case of labour, by a multiplicity of causes”—generates the regional specialization that is at the roots of international trade. Transport costs limit trade and, hence, the spreading of “cumulative upward and downward movements in money demand” (pp. 408–409). However, the degree to which transport costs curb transnational spreading depends on the income elasticity of import demand, which varies with the phases of the cycle. The longer a boom, “the more it is likely to ‘spill over’ into other regions,” due to bottlenecks in the domestic economy (p. 411).

Haberler subsumes the effects of tariffs and other “artificial” trade barriers under “the influence of transport costs” (ch. 12, § 2), since they too reduce the mobility of goods. Yet, while transport costs are primarily determined by physical geography and transportation technology, tariffs “can be altered by legislation, and therefore may be changed as a matter of policy systematically throughout the course of a cycle.… It is conceivable, therefore, that a country might use its power to vary its tariffs as a means of mitigating the violence of its own industrial fluctuations” (pp. 413–414). Haberler cautions that countercyclical trade policies at the national level tend to have procyclical effects at the global level, with negative feedbacks on the country of origin (p. 414).

The next international complication of business cycle theory is the influence of imperfections on the mobility of capital, which Haberler (ch. 12, § 3) describes as “localization of investment, credit and banking.” While the immobility of goods

Haberler argues that observable differences in interest rates indicate imperfect mobility of capital, for which his basic explanation is a pervasive home bias of investors. The barriers to cross-border lending “are less physical than political, social and institutional. Ignorance of foreign tongues, inadequacy of legal protection, risk of transfer restriction or outright confiscation … are nowadays certainly more important factors than actual distances” (pp. 417–418). The segmentation of financial markets mitigates local booms and crises, insofar as the supply of loanable funds is restricted and locked in. If, on the other hand, the boom is caused by a rise in goods exports, the inflow of revenues may depress interest rates below the levels that would have prevailed in open financial markets, thus amplifying the cyclical movements.

Apart from home bias and other barriers to mobility, capital flows have a tendency to vary in a procyclical fashion. As investor confidence falls, foreign investment is overproportionally reduced during the downswing phase of “the world business cycle.” “[T]he brunt of the depression falls on the borrowing countries, while countries which are normally capital-exporting obtain some relief through the improvement of their balance of payments” (p. 425).

Haberler suggests that the “rôle of capital movements in the course of the world business cycle … is largely determined by events on the foreign exchanges” (ibid.). Exchange rate policies are the final international complication in his synthetic exposition of business cycle theory (ch. 12, § 4). Haberler proceeds systematically “top down” through a range of real and ideal exchange rate regimes. He starts with the fiction of a global monetary union and full capital mobility in terms of a single “basic money” and “a single pool to which would-be borrowers from all parts of the system had equal access” (p. 426). This constellation does not exclude the occurrence of local booms and depressions, as the demand for funds might be concentrated in certain regions due to “a certain localisation of real resources.” Yet, monetary policy will neither block nor spread cyclical impulses between the regions.

Introducing further institutional features, Haberler switches to a strict gold standard with imperfect capital mobility, then to “exchange standards” (adjustable pegging) with exchange rate speculation, and finally to “free exchanges” (floating) with differing assumptions about capital movements (pp. 427–449). The decentralization sequence is described as an increase in national currency autonomy. This gives domestic monetary policy increasingly more power to counteract cyclical fluctuations, while also increasing the potential of speculative flows of capital. Variations in the assumptions about capital mobility, the monetary policy stance, fiscal and trade policies, etc., lead to a kaleidoscopic multitude of cases, in which the outcomes mutate considerably with the constellations of assumptions.

While an increase in monetary autonomy might be expected to reduce the international spreading of cycles, Haberler shows that there are some constellations in which autonomy contributes to convergence of the cyclical movements, while other constellations lead to divergence, such that the effects of capital flows on their destinations are diametrically opposed to the effects on the countries of origin. He arrives at the “somewhat paradoxical result that, under free exchanges [floating], capital imports and, under an international standard [pegging], capital exports are likely to have a deflationary influence” (p. 450). In the case of pegging, the tendency towards deflation is introduced by a reserve drain, whereas an increase in capital imports under floating is normally caused by an appreciation that tends to depress the sales of domestic products.

Table 1 illustrates Haberler’s systematic variation of assumptions for “different degrees of national currency autonomy” and the kaleidoscopic changes in expected outcomes. It should be noted that this is a strongly reduced version. Haberler’s taxonomy is more complex, permitting greater variation of outcomes by further differentiation of institutional features. To give an example, Haberler expands his analysis of the spreading of cycles in unified monetary systems from a monetary union with a supranational central bank to a single-currency system with a greater number of national central banks; the results differ, in particular with regard to the relative importance of fiscal policies. It is certainly all hypothetical, and at the end of the chapter Haberler warns the reader accordingly that his chapter on international aspects of business cycles is “containing an exposition of a method of analysis rather than a presentation of definite results” (p. 451). Yet, with hindsight on postwar developments in international monetary history, his selection of relevant scenarios appears as clairvoyant. With his kaleidoscopic method Haberler anticipated, if not influenced, the Mundell–Fleming approach to open economy IS-LM analysis that emerged in the 1960s and held sway over modern macroeconomics for a long time to come.

Neisser’s Theory of Cyclical Growth

Contrary to Haberler and “customary theoretical analysis of business cycles,” Neisser (Reference Neisser1936, p. 1) does not treat “the world as a unit.” He aims at developing the concepts of his business cycle theory in “the narrower boundaries of a national economy” before proceeding to analyze the international complications.

Neisser’s benchmark for business cycle analysis is a “moving equilibrium,” in which “receipts do not fall short of costs” (p. 4), including a normal return on capital. In the ideal case, this dynamic equilibrium is a steady-state growth path on which capital accumulation, net of increases in capital intensity, is strong enough to compensate for the labor-saving effects of technical progress. Full employment is preserved. In a series of better-known papers in the Kiel journal (1932), the Journal of Political Economy (1934a), and the American Economic Review (1942), Neisser argues that the growth process of a capitalist economy can be described “as a race between displacement of labor through technological progress and reabsorption through accumulation,” in which “displacement and accumulation are two largely independent factors” (1942, p. 70). Since the implementation of technical progress tends to require “capital intensification” (increases in capital intensity), the pace of accumulation is not necessarily sufficient to generate a capital stock that has the capacity to reabsorb displaced labor. Even when, “for the system as a whole, receipts do not fall short of costs, … equilibrium is not necessarily characterized by full employment.… Labor … might be unemployed in ‘equilibrium’, the real wages being the ‘prevailing ones’ (fixed by custom or union agreement) or already pressed down to the minimum level of subsistence” (1936, p. 4). Neisser (Reference Neisser1936, pp. 62–64) defines such persistent technological unemployment as “structural unemployment”; more specifically, it could be described as capital-shortage unemployment.

As Neisser shows in his 1936 book, structural unemployment may result from cyclical fluctuations of economic activity. Cycles start when the structures of production and expenditure fail to match in the course of technical progress and capital intensification. Receipts for sales may rise at first, but will fall short of costs, as endogenous contractions of credit lead to a general glut and the transformation of cyclical into structural unemployment.

Corresponding to his definition of a dynamic equilibrium, Neisser characterizes all shortfalls of receipts vis-à-vis costs as overproduction. They occur frequently in various sectors of a decentralized economy and amount with some regularity to partial overproduction in the sphere of producer goods (I-sphere) or in the sphere of consumer goods (C-sphere). Whether discrepancies between the structures of production and demand develop into general overproduction, and hence to a crisis, depends largely upon the sphere in which they arise; Table 2 illustrates the basic scenarios.

A case typical for a “growing system” is overproduction in the I-sphere. Neisser describes this as “undersaving” or “capital scarcity,” since savings from income in both spheres fail to fully finance expenditure on net investment. The mismatch of supply and demand for producer goods may be caused by technical progress (increasing their supply), by a sudden rise of the wage level (increasing the demand for consumer goods), or by “forced saving” through an overproportional rise in the prices of consumer goods (increasing the relative profitability of their supply), as in Hayek’s Prices and Production (1931).

Contrary to Hayek, however, Neisser argues that undersaving will not necessarily generate a crisis. ^{Footnote 6} Overproduction in producer goods industries tends to remain partial, as consumer goods industries profit from lowered costs of inputs. Increased demand lets profit expectations in the I-sphere recover and generate an upswing in total activity. The economy returns to dynamic equilibrium or overshoots into a boom. If, on the other hand, there is overproduction in the consumer goods industries—“oversaving” in Neisser’s terminology—it tends to become general, since it depresses the demand for capital goods. This affects profit expectations negatively in both the C- and I-spheres. “The peculiar feature of oversaving lies in the fact that it may immediately set into motion a mechanism rendering general the partial overproduction and creating the type of a general, self-intensifying depression well known at least since 1929” (1936, p. 88).

The discharge of bank debt in modern financial systems can quickly reduce the volume of money. If expectations of losses make this happen on a larger scale, partial overproduction—even if starting only in the I-sphere—sets off a vicious circle of contractions in the credit volume, losses of income, and deflation. In this case, receipts fall short of costs throughout. Partial overproduction is transformed into general overproduction and a recession that “is essentially a monetary phenomenon arising from absolute or relative deflation” (1936, p. 30).

The “working of the credit mechanism explains, also, why the crisis … must terminate sooner or later.” The repayment of bank loans plus hoarding “leads eventually to a condition of liquidity that almost forces the granting of new credit to those industries which have readjusted their cost or show profit for other reasons” (1934a, p. 446). The readjustment of costs includes investment in labor-saving devices that makes use of technical progress and precludes the reabsorption of workers who have been laid off in the crisis. A new credit expansion and cycle may start—though possibly at a higher level of unemployment, since the cyclical unemployment generated by the previous crisis is (at least partly) transformed into structural unemployment.

International Aspects of the Cycle

The international aspects in Neisser’s theory of business cycles are a complex assortment of factors that reinforce the effects of partial and general overproduction. His framework contains constellations of international relations that do not only transmit, but also generate, cyclical fluctuations. Those complications are deeply engrained in the structure of production and international division of labor.

Throughout the book, Neisser draws dividing lines between industrial countries on the one hand, and non-industrial countries on the other. He describes the latter as “foodstuff” and “raw stuff countries,” and frequently distinguishes between “colonial countries” and “new countries” (such as Argentina, Australia, and Canada). As Table 3 illustrates, the distinction between industrial and non-industrial countries is not fully congruent with the distinction between the I- and C-spheres. Industrial countries specialize in the production of producer goods and industrial consumer goods, while non-industrial countries deliver raw materials (I) and foodstuffs (C) (pp. 24–25). This interdependence reduces the risk of regional overproduction. International trade helps to match supplies and demands among different types of countries in the world economy (p. 30).

Neisser points out that there is no direct correspondence between international trade in capital goods and flows of financial capital (pp. 23–25). Foreign investment may finance domestic production and any trade, while imports of capital goods may be entirely financed by domestic sources. The international spreading of business cycles is not uniquely determined by capital flows. They could also contribute to a “localization” of cyclical fluctuations, but, nevertheless, they play a key role in Neisser’s analysis, starting from his observation that industrial countries are “usually” net savers and, hence, exporters of both capital goods and financial capital (1936, p. 46). Industrial countries are therefore prone to oversaving crises rather than undersaving—with the exception of Germany in the late 1920s.

As Neisser (Reference Neisser1936, pp. 51–61) explains, Germany suffered from capital scarcity from 1925 onwards. The crisis was detained by capital imports, mainly from the United States, but broke out in full when capital inflows stopped abruptly after the Wall Street crash in 1929. German undersaving was essentially caused by the imports of raw materials and foodstuffs required for the expansion of the I-sphere and outward transfers for war reparations. ^{Footnote 7} Both factors led to current account deficits, despite increasing exports. Overproduction of capital goods went hand in hand with labor saving by mechanizing “rationalization” in German industries, which created structural unemployment. “Despite huge capital imports, the capital stock increased too slowly to absorb in full the fresh supply of labor, created by displacement proper and by the natural growth of the population” (p. 68).

Structural unemployment is not confined to undersaving Germany in the 1920s. According to Neisser, it is a fundamental “dilemma of recent capitalism,” which most prominently applies to Britain, too (pp. 69–86). It results from a trend of capital intensification in the production of producer goods and durable consumer goods, and from the decreasing “permanent employment power” of consumer goods industries.

The “dilemma” is another version of the race between displacement of labor by technical progress and reabsorption by capital accumulation, now complicated by international exit options for investors. Negative short-run effects of capital outflows on domestic employment can be, but need not be, compensated by their long-run feedbacks to export demand:

Employment Effects of Capital Imports

Many of the “more profitable investments abroad” are made in non-industrial countries and help these “to complete the structure of domestic production by industrialization” (p. 50). Capital imports could thus avert undersaving even in colonial countries. Yet, Neisser (Reference Neisser1936) argues extensively, there are many systematic obstacles to international finance of growth and employment in developing countries. They contribute to cyclical effects of capital flows to non-industrial countries even when savings are growing steadily at the global level. Neisser demonstrates this in his analysis of the employment effects of international finance, considering “the limiting case of a colonial country that would spend the whole amount of foreign funds for importing capital goods” (p. 47). The absorption of unemployed workers would depend on “the employing power of new enterprises established by assembling the imported capital goods” (ibid.). The wage incomes of those enterprises create secondary employment in a multiplier process that continues “until the original amount of imported foreign funds is converted entirely to foreign goods, which, for any reason, could not be produced in time in the capital-importing country” (p. 48). Neisser derives the following multiplier expression (here written in symbols more conventional than those used by Neisser):

where l denotes additional employment, k the net imports of capital, y the net output per worker, and µ the import coefficient of spending; i.e. “the fraction of expenditure spent on the average for foreign products” (ibid.). Neisser does not consider the import coefficient µ as constant, but as endogenous to capital imports, while k may be reduced by the capital exports of “absentees” who take their savings out of the non-industrial countries (p. 25). Moreover, capital imports of colonial and “new countries” tend to carry greater risks. This makes them limited and volatile. The capital flows depend “upon unusual chances of return (exploitation of natural resources or land speculation … )” (p. 46). Booms in the capital-importing countries increase the imports of consumer goods, thus increasing the value of ì and lowering the local employment effect (p. 27).

The employing power of capital imports declines if y takes high values. Capital imports go primarily to “governments, municipalities, and other public agencies who internationally are preferred as borrowers” (p. 50). They invest predominantly in “public utilities, railroads, and buildings,” which show high employment effects during their construction period, but much less during utilization. Due to the volatility of capital flows to colonial countries, Neisser finds investment there to “be subjected to heavy fluctuations even if the flow of savings in the world as a whole displayed a steady course” (p. 46). The redistribution of capital flows at the international level generates cyclical impulses in the non-industrial countries.

The International Spreading of Depressions

The problems that capital flows and their reversals cause in non-industrial countries become even greater when oversaving and depression occur in the industrial countries. In the general framework of his theory, Neisser (1936, pp. 88–90) identifies the following causes:

1. (i) reduction of consumption by “displacement of workers by labour-saving devices”;

2. (ii) “‘forced saving’” by “one-sided investment in consumer goods industries”; and

3. (iii) wage stagnation in the face of total output expansion.

According to Neisser, a mix of all these causes led to the outbreak of the Great Depression in the United States in 1929. It provoked a sudden change in profit expectations that prompted the crash in the stock market (pp. 90–99).

When oversaving turns into depression, price and wage levels tend to fall. A combination of hoarding and liquidation of capital intensifies the deflation. As Neisser points out, the deflationary impulses from a “principal center,” such as the US, are propagated to the rest of the world economy essentially through three channels (pp. 106–115):

1. (1) reversals and contractions of capital flows (withdrawals of loans and debt collection);

2. (2) reductions in imports of all types of goods; and

3. (3) downward pressure on “the international price level” due to the shrinkage of the domestic market and intensification of competition in the world market.

Table 4 provides an illustration of these channels of transmission and the sequence in which they are typically activated. Reversals of international capital flows can make oversaving in one region coincide with undersaving in another; they can even generate undersaving in the other region. Reduced access to international liquidity bars the elimination of complementary excess supplies. Neisser demonstrates this by referring to simultaneous oversaving in the US and undersaving in Germany, when capital flows broke down in 1930: Germany had to reduce both domestic consumption and capital goods exports (pp. 99–101). The more general case is simultaneous oversaving in industrial countries and undersaving in non-industrial countries that finance production in raw material or food sectors largely with external debt.

In the process of debt collection, debtor countries lose reserve assets and face the choice between deflation and an autonomous increase of their money supply. The latter requires exchange controls or exchange rate depreciation (1936, pp. 107–109). If depreciation leads to a rise in commodity exports, it could reverse the reserve losses. In a worldwide depression it will be difficult, however, to compensate the reversal of capital flows by increases in goods exports, as demand from creditor countries is likely to fall short. In general, debtor countries are forced to achieve a “sudden and rapid liquidation of debts” through fire sales. Their externally financed production is critically exposed as undersaving and overproduction in the I-sphere. ^{Footnote 8} The panic sales become “an immediately acting factor in a world depression,” since they induce capital flight and imbalances in the structures of production in debtor countries.

Mechanism (1), the collection of debt, is limited by the volume of international lending, and may not last long. But mechanisms (2) and (3), the decline in goods imports of central economies and the intensified competition in shrinking markets, are most likely “to frustrate the efforts of the surrounding world to isolate the depression.” It is more dangerous because “a discrepancy between saving and investment, as implied by the third mechanism may last for an indefinite period” (p. 110). The deflation imposed by the central economy could spur “autonomous deflation” in other countries “when, due to the international deflationary pressure, profits fall below the critical mark, creating a discrepancy between savings and investment and even deflationary anticipations” (p. 111). Autonomous deflation is brought about by an increase in hoarding, indicated by a decline in the velocity of circulation of money. ^{Footnote 9}

Autonomous deflation generates “secondary unemployment” and provides a motive for protectionist measures to fight the depression. The reabsorption of displaced workers tends to be confined to parts of the economy with low marginal productivity of labor. Competition in the labor market is then likely to induce “a continuous fall in the [real] wage level.”

In his discussion of “resistance to deflation,” Neisser (Reference Neisser1936, pp. 112–115) observes a cyclical decline in the terms of trade of countries that export primarily food and raw materials. ^{Footnote 10} He explains this with differences in market structures. “In agriculture, competition is perfect insofar as the produce is destined for the world market” (p. 113). Agricultural producers are price takers, and prices in global food markets are more flexible than those in markets for manufactured goods. This makes the terms of trade rise for industrial countries that can uphold their living standards better than non-industrial countries—at least for the employed part of their population. Unemployment, on the other hand, may rise faster in the industrial countries. In order to avoid losses, price-setting manufacturers cut back on their output rather than on prices.

Finally, Neisser discusses “Some International Aspects of Recovery” (pp. 134–160), starting with an increase of liquidity in the banking system as a precondition. Analyzing the effects of interest-rate reductions, cost cutting, and fiscal stimuli, Neisser emphasizes that the “particular complicated set of conditions” for recovery must be considered in all theoretical work and policy advice.

This warning of a fallacy of composition does not imply advocacy of fixed exchange rates. In Neisser’s view, there is no clear-cut argument in favor of one type of exchange rate regime or another. The book ends on the somewhat frustrating note that “[t]here is no hope for simple remedies—as little as for simple explanations. Again, we cannot help feeling the need for more complicated economics” (p. 160).

The Representative MOEM

In the context of this paper the appropriate MOEM reference is Uribe and Schmitt-Grohé (Reference Uribe and Schmitt-Grohé2015), a textbook in the making, written by an Argentinian and a German, both trained at Chicago and now professors at Columbia University. Their draft (currently accessible online) is here considered as most representative for comparison with the approaches of Haberler and Neisser because it deals with open economies from a business cyle perspective.

In the first chapter, Uribe and Schmitt-Grohé (Reference Uribe and Schmitt-Grohé2015) present stylized “business-cycle facts around the world” for the period from 1960 to 2010. Many of their findings accord with the observations made by Haberler and Neisser in the 1930s; for example:

1. (1) Investment is more volatile than total output and consumption (Table 1.1, p. 7); compare with Haberler’s “acceleration principle.”

2. (2) The trade balance is countercyclical (p. 9); compare with Haberler’s argument about countercyclical changes in the elasticity of supply that raise import demand in the boom and export capacity in the depression.

3. (3) The cross-country average volatility of output is twice as large as its US counterpart (p. 7); compare with Neisser’s conjecture about the international spreading of depressions from the center(s) of the world economy.

4. (4) Business cycles in rich countries are about half as volatile as cycles in emerging markets and poor countries (pp. 11–12); compare with Neisser’s dualistic description of cycles in industrial and non-industrial countries, in which the latter suffer extra complications. ^{Footnote 13}

How do Uribe and Schmitt-Grohé (Reference Uribe and Schmitt-Grohé2015) explain their stylized facts? They develop a real-business-cycle model for the small open economy (SOE-RBC), into which they introduce a number of shocks (on interest rates, the terms of trade, public spending, etc.), in addition to the generic technology shocks. This enables them to explain facts (1) and (2). Thereafter, they deal with the potential of exchange rate policy to mitigate business cycles and reduce involuntary unemployment that arises from downward rigidities of nominal wages. The framework is thus expanded from SOE-RBC to a DSGE model with nominal rigidities. Unemployment is considered as an externality induced by currency pegs, and devaluations are shown to be welfare-improving. Taxes, subsidies, and capital controls are analyzed as first- and second-best fiscal policies to deal with shocks under fixed exchange rates. In the final chapters the focus is set on financial frictions in terms of lacking policy credibility and externalities that arise in the form of collateral constraints (depending on variables not controlled by individual borrowers, but determined by their collective behavior). Various models in these chapters help to explain the stylized facts (3) and (4).

The Uribe and Schmitt-Grohé manuscript connects aspects of international trade, finance, and growth in a unified framework of business cycle analysis. At first glance, it seems close to fulfilling Haberler’s and Neisser’s dreams of linking the theories of international trade and “economic change” by “more complicated economics.” By comparison, Neisser’s and Haberler’s approaches may appear antediluvian, looking like primitive explorations of an unpenetrated field—in the case of Haberler, entirely pedestrian, in essayistic prose; and in the case of Neisser, convoluted and rudimentary, despite some formal analysis and empirical evidence.

Reversing the Perspective

There is no denying that modern macroeconomics is far more sophisticated than the interwar literature, owing to the ongoing evolution of the world economy, vastly increased inputs in economic research, advances in modeling technology, and many other factors. But what if we look at MOEM from the perspective of Neisser and Haberler? Primitive art works often have their charms in making essential traits protrude in a simpler fashion, easier to grasp than the intricacies of modernity.

An important difference between MOEM and the two older approaches is their framing. In the SOE-RBC framework, the international aspects of the business cycle are analyzed from the perspective of a single and, in general, small open economy. It is the perspective of an island adapting to external shocks. Haberler, by contrast, started from an integrated world economy and discussed the properties of disintegrating factors that would modify the propagation of cycles between regions. Neisser started from cyclical disturbances in single economies of different types and proceeded to analyze their interdependence in the world economy. By comparison with the world economy perspectives of Haberler and Neisser, the single economy perspective of the SOE-RBC framework makes it more difficult to analyze the systemic risks of cyclical fluctuations.

Note that the split between the analytical perspectives is not about the requirement of microeconomic foundations. Neisser (Reference Neisser1936, p. 4) actually emphasized the need for them:

In his world economy perspective, Neisser set the focus primarily on profit expectations and out-of-equilibrium constraints of income and consumption. This differs fundamentally from the DSGE approach of dealing with observable cyclical fluctuations as if they were results of continuous intertemporal optimization of representative households. Neisser’s basic argument about tendencies of undersaving in debtor countries, oversaving in creditor countries, and critical reversals of capital flows provides a more straightforward approach to thinking about the macroeconomic imbalances that emerged prior to the recent global financial crisis.

A fundamental difference between MOEM and the approaches of Neisser and Haberler lies in their views of the driving forces of business cycles. In the DSGE framework the system moves along its steady-state growth path, unless it is subject to positive or negative shocks. These exogenous impulses lead to fluctuations that are interpreted as readjustments towards the steady state, unless disturbed by further shocks. The “cycle” is thus described in terms of responses of output, investment, and other endogenous variables that would not exhibit any cyclical behavior otherwise, since the system is well coordinated by rational expectations. Neisser and Haberler, on the other hand, stress the importance of “organic counterforces” that tend to make cyclical movements endogenous. Structural maladjustments lead to bottlenecks and partial overproduction that make the system vulnerable to “accidental disturbances” and crises. The latter are overcome when idle capacities and hoards of liquidity make the system sensitive to “stimulating influences.” The “shocks” that account for the turning points play a small role—they are accidental rather than decisive. The basic explanation for the cyclicality of growth is imperfect information about the investment required to match the structure of production with demand. Responses to errors produce new cycles rather than returning the system to the steady state.