Theories of Economic Crises

The theoretical approaches to analyzing crises have behind them contrasting conceptions of the way the economy works

1. Introduction

There are different understandings of economic crises. Distinguishing between them is essential for an orderly debate, both for the interpretation of crisis situations and for making judgments about possible economic policy interventions.

In the following, I will attempt to briefly sort out the main conceptions of crises: as a normal phase of the economic cycle; as an unexpected shock that moves (transiently) away from an equilibrium position, considered as the center of attraction of the economic system; as the result of the systemic instability of market economies, in particular of the game of financial expectations. The first two conceptions - the first two groups of theories - consider crises as accidents of the road with no lasting effects on the economy, whose long-term course depends essentially on developments in technology and resources, including the working-age population. The third conception - the third group of theories - sees crises as manifestations of the endogenous instability of market economies with persistent negative effects on income and employment trends, social coexistence and civil life, and the environment.

The theoretical approaches to analysing crises have behind them contrasting conceptions of the way the economy works, present both in the history of economic thought and in contemporary debate – despite the attempts of dominant mainstream views to silence the other. On the one hand, we have the marginalist conception of the economy, as a theory of rational choice between alternative uses of scarce resources, with all its variants; this includes the first two lines of analysis of crises, as oscillations around a long-term trend and as a transitory deviation from a position of equilibrium. On the other hand, we have the conception of the ‘classical’ economists of the eighteenth and nineteenth centuries of a circular flow of production, consumption, and exchange: a conception from which is absent the reference, characteristic of the marginalist approach, to equilibrium levels of production and employment corresponding to the full use of the productive forces, and above all the thesis of an automatic convergence towards such equilibria guaranteed by the play of supply and demand in perfectly competitive markets. In the classical conception, we can also include Keynesian theory, if correctly interpreted in its characteristic elements, and its developments concerning the systemic instability of market economies with a high degree of financialization.

2. Crises as normal phases of the economic cycle

The conception of the crisis as a normal phase of the business cycle, followed by depression, recovery, and boom dominated in the first decades after World War II, during what has been called the Golden Age, characterized by a more rapid development than in the preceding decades (the interwar period marked by the Great Crisis) or in the following ones (the last half century). During that period, the so-called neoclassical synthesis prevailed in the debate on economic theory: a compromise between the foundations of marginalist value theory and a tamed version of Keynesian theory. Of the marginalist theory, this compromise preserves the thesis of a tendency towards full employment equilibrium in the long run; Keynesian theory, reinterpreted as limited to the short run, is used to argue for the usefulness of active fiscal and monetary policies, aimed at countering at least the most extreme manifestations of temporary economic imbalances: expansionary policies to counteract unemployment in crisis and depression phases, and restrictive policies to counteract inflation in recovery and boom phases.

A typical example of this group of business cycle theories is the one proposed by Paul Samuelson, based on the interaction between accelerator and multiplier; but there are several others. These theories consist of mathematical models based on differential or finite-difference equations linking the development of income to that of investment, and the development of investment to changes in income (or expectations of such changes). An appropriate value of the parameters gives rise to a succession of growth, boom, bust, and trough phases (while other values of the parameters give rise to explosive trends or trends toward stationarity). A dense series of exercises then complicated the basic model by introducing taxation and money, international trade and income distribution, and other phenomena to show the limits and potential of active fiscal and monetary policies. The anchorage to marginalist theories of value is given by the fact that fluctuations take place around a long-run equilibrium path corresponding to a normal degree of utilization of available resources, including labor power (hence full employment). According to marginalist theories, in fact, changes in wages (the price of labor power) lead, as for all goods, to equilibrium between supply and demand in the labor market, on the sole condition that this is characterized by perfect competition.

The idea of crises as transient phenomena destined to automatically yield to full resource utilization is present, in the second half of the 19th century and the first half of the 20th century, among Austrian and Swedish economists. An original variant of this tradition is the theory of the business cycle proposed by Schumpeter in the first half of the last century and taken up several times in the most recent debates, which combines the theory of the cycle with that of development. According to Schumpeter, innovations continually upset the static equilibrium characterized by full employment. In addition, innovations tend to appear not as a regular flow but in swarms, thus initiating a phase of growth, generated by the investments of the innovating companies; these take resources away from the traditional companies thanks to the financing available to them and to the inflation produced by a demand that exceeds the availability of resources. When the additional output of innovative firms arrives on the market, price growth turns into a decline; traditional firms are gradually forced out of the market; a crisis and depression phase ensues. Market adjustment mechanisms return to an equilibrium of full utilization of available resources, but with higher levels of output and per capita income, thanks to innovations. This theory reiterates the thesis of the usefulness of crises, which release the resources needed for the development of innovative firms (the ‘creative destruction’ thesis). Obviously, for the validity of this thesis, it is necessary for the assumption to hold that market mechanisms ensure convergence towards equilibria of full use of resources. The same condition must have applied to those who, like Marshall, saw crises as a necessary purge of the speculative excesses of boom phases.

In the face of these positions, the subsequent compromise of the neoclassical synthesis, according to which active monetary and fiscal policies are useful to stabilize economic developments, while it is necessary to rely on market mechanisms to ensure long-run equilibrium, seemed to work in practice in the quarter century after the end of the Second World War. Then it went into crisis with the collapse of the Bretton Woods system and the phase of simultaneous high inflation and unemployment that followed the oil crises of the 1970s. The same period saw the rise of neoliberalism in political and cultural terms, monetarism, and then the theory of rational expectations in economic theory.

3. Crises as unforeseen shocks

Let us now consider the conception of crises as unexpected shocks that take the economy out of the equilibrium position, towards which market forces will tend to bring it back. The theoretical foundation of this conception is to be found in the theories of rational expectations, which carry to logical consequences the theses proposed in analytically more rudimentary forms by Friedman’s monetarism. According to these theories, economic agents, perfectly rational and endowed with complete information, take into account every element that exerts a systematic action on the economy. Only random elements, which can cause stochastic deviations from the equilibrium position, are excluded. Net of these stochastic deviations, the system is always in equilibrium: an equilibrium in which the real variables, income and employment, are not influenced by monetary and financial phenomena.

As mentioned above, crises resulting from unforeseen shocks remain possible: for example, an earthquake, a pandemic, but also erroneous economic policy interventions, or errors in the management of large enterprises that are driven to bankruptcy, or sudden financial crises, which can have disastrous consequences. Shocks of this kind can produce a fall in production and employment levels or, conversely, inflationary explosions, as happened during the successive oil crises, which were attributed to political events unforeseeable to economic agents. In each of these cases, the mechanisms of a competitive market then bring the system back into equilibrium.

Proponents of this group of theories are against any kind of government intervention in the economy. Counter-cyclical fiscal and monetary policies accepted by the neoclassical synthesis are considered unnecessary, since they operate in a systematic way and can be predicted by rational economic agents who take them into account in their decisions, thus cancelling out their effects. To proponents of this view, the difficulty of precisely regulating the course of the economic cycle, correctly forecasting its development and adopting the appropriate fiscal and monetary interventions at the required time and to the exact extent, makes errors in the management of economic policy likely: this creates shocks that take the economy off its equilibrium path.

If one starts from the traditional marginalist theory of value and distribution, these conclusions follow very logically. Faced with concrete economic events, it is always possible to find ad hoc justifications for crises, in one or another type of shock. Thus, for example, in the face of the financial crisis of 2007-2008, it was argued that it was the introduction of fiscal measures that caused an increase in voluntary unemployment that caused it; an article supporting this thesis was accepted by one of the leading journals in our profession.[1]

The problem with all these theories is that they are based on the automatic adjustment mechanisms towards full employment of marginalist theory, rooted in downward changes in wages in the face of unemployment and upward changes in labor demand in the face of a reduction in wages. This inverse relationship between real wages and employment has been criticized from different perspectives. Keynes pointed out that the prospects of falling demand linked to a reduction in the real wage put downward pressure on consumption and investment, and thus on aggregate demand, output, and employment.[2] Sraffa showed that a reduction in the real wage does not necessarily make the use of more labor-intensive techniques cheaper. A large subsequent debate confirmed the validity of this criticism. General economic equilibrium theories themselves have come to the conclusion that full employment equilibria, in addition to being multiple, can also be unstable.

4. Crises and systemic instability in market economies

Keynesian theory is based on a sequence of cause-and-effect relationships: what happens in the money and financial markets, over which expectations dominate, in a very short-term perspective that favors instability, affects interest rates and more generally the conditions under which investments can be financed. Decisions on the latter are taken from a long-term perspective, but the timing of realization can then be adapted to the evolution of the financial markets and to expectations on sectoral and aggregate demand. As a result, investments fluctuate over time, both because expectations of returns change and because it is more or less easy or expensive to obtain the necessary financing; this in turn leads (via the multiplier mechanism) to fluctuations in output and employment. Moreover, there is no reason to assume that these fluctuations occur around any equilibrium level or trend towards full employment.

We thus have two implications of Keynesian theory for the conception of crises: the crisis as a depression, i.e., as the persistence over time of even high levels of unemployment; the crisis as instability, episodes of falling employment.[3] The economic policies prescribed therefore consist of both systematic support for levels of aggregate demand and interventions to reduce instability. In both cases, it is not just a matter of adopting expansive or restrictive monetary and fiscal policies, but of creating an environment of rules and customs conducive to the development of the economy. To give a few examples: control of speculative financial activities (as in Keynes’ hostility to short-term international capital movements), reduction of uncertainties in national economies and international relations (as in Keynes’ choice of fixed exchange rates at Bretton Woods), public investment policies in infrastructure and the environment, support for public education and widespread welfare.

After Keynes, the theory of systemic financial instability was developed in particular by Hyman Minsky. He distinguishes three types of positions adopted by economic agents: hedged positions, in which it is expected that the debt service with which the purchase is made of real or financial assets (e.g. houses, raw materials, machinery, shares, bonds) is more than covered by the expected income; speculative positions, where my expected income is higher than the repayment instalments on the loan under normal conditions, but may not be so if conditions change for the worse (e.g. when a company invests in machinery, which it would not be able to repay if sales of the product collapsed, or when the investment is financed with short-term debt that has to be refinanced, and I am then faced with a credit crunch); ultra-speculative positions, in which one or a few future events are crucial for servicing the debt (e.g. if I use loans to buy an asset, such as gold or silver, that yields nothing, I am betting everything on the fact that its price will rise over time at a rate higher than the interest rate). The ‘financial fragility’ of an economic system depends on the proportion between the three types of positions: it is higher the more widespread are positions of the second and especially the third type. With ultra-speculative operations, all it takes is a reversal in the price of the asset, or an increase in interest rates, to determine the bankruptcy of the operator involved, and if operations of this type are very widespread, there can be a general crisis of the economy. This was the case in 2007-08 when house prices stopped rising, causing a crisis for those who had bought them by taking out mortgages whose instalments were paid at least in part by taking out new loans, guaranteed by the increase in the price of the houses themselves; in the wake of the operators with ultra-speculative positions, financial institutions, large and small, which had mortgages that were now in default, went into crisis.

According to Minsky, crises of this type have a repetitive pattern, gradually increasing in intensity. Confronted with a crisis, political authorities adopt rescue measures. When the economy recovers, reassured by public intervention, financial operators start to build up ultra-speculative positions again: they guarantee huge profits if things go well, while the belief spreads that if things go badly, it will be the authorities who will get everyone out of trouble. Thus, from cycle to cycle, crises become heavier, while the interventions of the monetary authorities become more and more substantial until we pass from a situation in which financial institutions are too big to be allowed to fail to a situation in which they become too big to be rescued (from ‘too big to fail’ we pass to ‘too big to be rescued’: what happened in the crisis that started in 2007-8, fortunately only in the case of the Icelandic banks). We are therefore faced with the possibility that sooner or later there will be a crisis of colossal dimensions, such as to determine a real collapse of the world economy.

(The idea of a terminal collapse of capitalist economies recalls, but on a different basis, Marx’s theses, which cannot be discussed here, but which on critical analysis also turn out to be based on an erroneous labor-value theory).

5. Crises: opportunity or calamity?

As we have seen, according to Schumpeter’s theory, taken up in partially different forms in the more recent theory of expansive austerity, crises are not only an inevitable phase of the economic cycle but also a necessary phase for economic development. In fact, the bankruptcy of the least efficient firms that occurs in the crisis phase is necessary to free up the resources used by firms that invest in introducing innovations, and thus realize technical progress. Crises are therefore an opportunity to help the economic system get rid of the dross - the less efficient firms - and move forward.

This thesis, however, is only correct if the marginalist approach, into which Schumpeterian theory falls, is correct. For only in this case does the economic system automatically tend towards the full utilization of available resources, so that what entrepreneurs invest must necessarily be taken away from someone else. But this is a mistaken assumption, as we have seen, without which the sacrifices imposed by the crisis are useless if not counterproductive.

There is, in fact, a substantial difference between the two types of crises. On the one hand, both the crises theorized by the neoclassical synthesis, as phases of an oscillatory trend around an equilibrium path, and those considered the effect of unexpected and transitory shocks by the theory of rational expectations, are seen as episodes, all in all compensated by phases of recovery in the long-term trend of an economic system that grows in line with the available resources, therefore at the maximum possible rate. On the other hand, both Keynesian crises and the financial crises theorized by Minsky are instead downward deviations from the levels of full use of resources and full employment, towards which the system does not automatically tend: it is therefore a matter of a pure loss of production and employment.

Not only this: the presence of links between production levels and growth rates on the one hand, and technical progress on the other (static and dynamic economies of scale, learning by doing, etc.) mean that crises also entail a loss of technical progress, which is not recovered over time, and thus move the economic system onto lower growth paths than would have occurred in their absence.[4]

Social cohesion can also be challenged by high levels of unemployment, income losses, and even uncertainty about job and income security. There has been a long debate, particularly in the 18th century, between those who argued that poverty and deprivation stimulate active reactions, which are also sources of improvement for the economy as a whole, and those who argued that both labor skills and ingenuity are negatively affected by conditions of deprivation and economic insecurity; with Adam Smith, the latter position has prevailed, confirmed by various empirical works in more recent decades. Moreover, the spread of education, which is increasingly important for economic and civic development, is correlated with income and employment levels.

Even in the face of the - extremely serious - environmental problem, unless we follow the path of degrowth, which is unlikely to be ‘happy’, or want to reach a situation of ecosystem collapse, the path of sustainable development requires technological changes, investment in infrastructure and regulation of production techniques which, in addition to requiring the state to drive and direct it, have a cost and are therefore more easily achievable in a situation of good economic performance.[5]

All in all, it can be said that economic crises are not an opportunity, but a great trouble for our societies. The opposite can only be argued on the basis of economic theories that are wrong in their foundations. The economic policy choices made on the basis of such theories have in fact led to serious and growing problems; a cultural renewal in the debate on economic theory is needed to meet the serious challenges on the horizon.


[1] L.E. Ohanian, “The economic crisis from a neoclassical perspective,” Journal of Economic Perspectives, vol. 24, 2010, pp. 45-66. Again, in order to avoid acknowledging what are in fact normal market dynamics, the entire responsibility for crises has been attributed from time to time to errors or corruption on the part of regulators (which do exist and can increase problems, but do not cause them).

[2] According to rational expectations theory, on the other hand, the fall in consumption caused by the reduction in real wages would be accompanied by an increase in investment, because more capital-intensive techniques would become cheaper. However, it was precisely the inverse relationship between real wages and capital intensity of techniques that was conclusively criticized in the 1960s debates on capital theories. On these issues and the surrounding debates see A. Roncaglia, The Age of Fragmentation, CUP, Cambridge 2019.

[3] Keynes argues the first point mainly in the General Theory, the second mainly in earlier and later works (see M. Tonveronachi, J.M. Keynes. Dall’instabilità ciclica all’equilibrio di sottoccupazione, NIS, Rome 1983).

[4] Some exponents of the neoclassical synthesis have pointed to something similar to the phenomenon of hysteresis, whereby long-run equilibria are modified by short-term fluctuations. In this way, they emphasize the importance of active policies to counteract crisis and depression phases of the cycle, while retaining the basic reference to a persistent tendency towards resource-full equilibria, which is the subject of the criticism mentioned above.

[5] The sustainable development theses referred to here are distinct from the neo-Malthusian theses of the Club of Rome, which were based on the thesis of a scarcity of natural resources (it was predicted in the early 1970s that oil would run out within 18 years). As with Jevons’ earlier thesis, according to which British manufacturing development would be halted by the depletion of coal, these theses do not take into account the effects of technical progress, which in the long run have proved decisive, and whose active exploitation is at the heart of the sustainable development thesis.


Alessandro Roncaglia is Emeritus Professor of Economics at La Sapienza University in Rome. He is a member of the Accademia Nazionale dei Lincei in Rome and the author of many books and articles. His Power and inequality: A Reformist Perspective, will shortly appear as part of INET’s book series with Cambridge University Press (Italian edition: Il potere, Laterza 2023). This essay, published in INET’s website (, is adapted from a presentation at a conference at the Lincei that will be published in Italian later this year.

Alessandro Roncagllia

Professore ordinario di Economia politica, Facoltà di scienze statistiche, Università di Roma I