Thursday, December 03, 2009

The Economics of Hyman Minsky

There has been a resurgence of interest in the economic writings of Hyman Minsky over the past few years, and for good reason. I find this immensely gratifying. I first came across his work as a graduate student, and remember being struck by the manner in which he was able to weave together sweeping macroeconomic hypotheses with rich institutional and historical detail. In a 2002 review of a collected volume in his honor, I summarized the broad outlines of his argument as follows:
Minsky's theoretical framework combines a cash-flow approach to investment with a theory of financial instability. Investment is motivated by the expectation that it will yield a stream of cash flows sufficient to cover contractual debt obligations as they come due, while allowing for adequate margins of safety. During stable expansions, profits accrue disproportionately to firms with the most aggressive financial practices, resulting in an erosion of margins of safety. This raises the probability that a major default will trigger a widespread crisis. When a crisis does eventually occur, its most devastating impact is on the highly indebted firms that prospered during the expansion. Balance sheets are purged of debt, margins of safety rise, and the stage is set for the process to begin anew. From this perspective, expectations of financial tranquility are self-falsifying. Stability, as Minsky liked to put it, is itself destabilizing. The real effects of a crisis depend on the size of government and the role of the central bank. Large countercyclical budget deficits can sustain profit flows even when investment collapses, preventing a crisis from resulting in a debt-deflation. Similarly, lender-of-last-resort interventions can keep a crisis from spreading beyond its point of origin.
Ten years earlier I had grappled with the question of whether or not Minsky's financial instability hypothesis requires substantial departures from economic rationality at the level of individuals and firms. I tried to argue in a 1992 paper that it did not, because even sophisticated models of learning at the individual level need not result in convergence to rational expectations equilibrium paths. The formal analysis in that paper is largely forgettable, but it does have the merit of containing the following reasonably faithful overview of Minsky's conceptual framework:
An essential feature of Minsky's financial instability hypothesis is that a long period of sustained stability gives rise to changes in financial practices which are not conducive to the persistence of stable growth:
"The natural starting place for analyzing the relation between debt and income is to take an economy with a cyclical past that is now doing well. The inherited debt reflects the history of the economy, which includes a period in the not too distant past in which the economy did not do well. Acceptable liability structures are based upon some margin of safety, so that expected cash flows, even in periods when the economy is not doing well, will cover contractual debt payments. As the period over which the economy does well lengthens, two things become evident in board rooms. Existing debts are easily validated and units that were heavily in debt prospered; it paid to lever. After the event it becomes apparent that the margins of safety built into debt structures were too great. As a result, over a period in which the economy does well, views about acceptable debt structure change. In the deal making that goes on between banks, investment bankers, and businessmen, the acceptable amount of debt to use in financing various types of activity and positions increases. This increase in the weight of debt financing raises the market price for capital assets and increases investment. As this continues, the economy is transformed into a boom economy." (Minsky, 1982, pp. 65 f.)
In particular, there is an increase in the degree to which investment is debt financed, and in the ratio of "speculative" to "hedge" financing. Speculative financing differs from hedge financing in that the expected cash receipts fall short of the debt-repayment commitments in some periods, particularly in periods which are in the immediate future with respect to the date at which the investment is undertaken. Hence there arises a need to roll-over debt during initial periods after the investment. Since the present value of all expected future cash flows exceeds the present value of debt repayments by an amount which includes acceptable margins of safety, firms do not generally foresee problems with regard to raising funds for purposes of rolling-over debt during the initial periods. Nevertheless, the possibility of present value reversals in the face of sharp rises in interest rates cannot be ruled out entirely, and this is what makes speculative financing riskier from the point of view of economy-wide-stability. On the other hand, if firms were able to borrow an amount which was strictly constrained by their expected receipts in every period, the potential for expansion would be severely curtailed and several reasonable profit opportunities would have to be foregone. The degree to which bankers are willing to engage in speculative finance clearly depends on their state of confidence with regard to general economic conditions in the future, as well as the merits of the specific projects at hand. The willingness to refinance positions as debt repayments become due is likewise influenced by their expectations, as well as the liquidity constraints they face.
Hence hedge financing is conducive to stability while speculative financing is conducive to faster expansion. A sustained period of stability gives rise to optimistic expectations and a rise in speculative financing. As long as the investments undertaken during this period are effective in raising productivity and engender a steady expansion of output, the optimism can be sustained. However, if a large number of investments which are prompted by the availability of speculative finance are found to be inept, so that immediate cash flows are significantly lower than expected, then the need for short-term refinancing becomes acute while at the same time banks are less willing to roll over existing debt. A sharp rise in short-term interest rates occurs which can lead to present value reversals, a rush towards liquidity, a plunge in the prices of illiquid assets, both real and financial, and a corresponding drop in new investments. Such an event, depending on its severity, is generally described as a credit crunch, a state of financial distress, or a financial crisis.
The severity of the recession that follows on the heels of a financial crisis depends on the size of the government relative to the private sector and the degree to which the drop in investment is compensated for by a rise in the government deficit. A sufficient rise in the deficit can help maintain the gross profits of business and hence validate the existing liability structure. This is achieved, however, at the cost of sustaining an inefficient industrial structure and inevitably leads to accelerating inflation. In the case of small government or a government that is determined to maintain a balanced budget, the financial crisis is likely to be much more severe, and the collapse of gross profits, by further curtailing investment, may be expected to give rise to a protracted depression. Lender-of-last-resort action by the monetary authorities can help offset the liquidity squeeze, preventing an extreme escalation of short-term interest rates and allowing the refinancing of positions which have been created through speculative financing. The fall in asset prices which is engendered by the rush towards liquid assets can thereby be contained, so that the financial crisis is less acute. There is a serious long-run cost to indiscriminate lender-of-last-resort action, however, in that the expectation of such action on the part of banks and firms, once it has become established, leads to the general (and correct) perception that the risks associated with speculative financing are not too severe. Imprudent financing may thereby be encouraged.
To summarize, there are inherent tendencies in advanced capitalist economies with sophisticated financial structures which give rise to instability in financial markets. The severity of the financial crises that occur, and the subsequent losses in output and employment depend on the size and actions of the government. Instability is generated endogenously, in the sense that a period of sustained stability encourages an increase in speculative financing which renders the financial structure fragile. Instability on the other hand encourages financial prudence, and thereby creates the conditions for a period of sustained growth. It is within the context of such dynamic interactions that Minsky is able to speak of the "destabilizing effects of stability."
Expectations play a central role in the precipitation of the crisis, as well as in the creation of the conditions which characterize financial fragility. The basic assumption made by Minsky with regard to the formation of expectations is reflected in the following passages:
"Financing is often based on an assumption 'that the existing state of affairs will continue indefinitely' (GT, p. 152), but of course this assumption proves false. During a boom the existing state is the boom with its accompanying capital gains and asset revaluations. During both a debt-deflation and a stagnant recession the same conventional assumption of the present always ruling is made; the guiding wisdom is that debts are to be avoided, for debts lead to disaster." (Minsky, 1975, p. 128)
"A recovery starts with strong memories of the penalty extracted because of exposed liability positions during the debt-deflation and with liability structures that have been purged of debt. However, success breeds daring, and over time the memory of the past disaster is eroded. Stability -- even of an expansion -- is destabilizing in that more adventuresome financing of investment pays off to the leaders, and others follow." (Minsky, 1975, p. 126)
Although fully conscious of operating in a cyclical economy, the individuals described above are prone to making decisions, particularly with regard to liability structures, on the basis of assumptions that are periodically discovered to be false. This exposes Minsky's analysis to the charge that his explanation of financial instability depends on an assumption of irrationality on the part of financiers and entrepreneurs, in the sense that they persist in holding expectations that are systematically wrong.
But is it really true that Minsky's theory depends on an assumption of significant and persistent departures from economic rationality on the part of firms? I tried to argue that this was not necessarily the case:
From a theoretical point of view, it is desirable to determine what manner of departures from rationality, if any, are required for Minsky's central propositions to hold. Bernanke (1983, p. 258) has stated for instance that Minsky and Kindleberger "have in several place argued for the inherent instability of the financial system, but in doing so have had to depart from the assumption of rational behavior." Bernanke does not elaborate on the precise nature of the departures which are a necessary component of the FIH, and his conclusion is somewhat perplexing in the light of Minsky's repeated assertions to the effect that "capitalism abhors unexploited profit opportunities" (1986, p. 219). In fact profit maximizing banks and firms are central actors in the scenario that is depicted, to the extent that financial innovation itself is explained on the basis of self-interested behavior, often in response to policy initiatives. It is possible that Bernanke is referring to irrationality with regard to expectation formation, and if this is the case, then his sentiments are shared by Flemming who believes that "the argument depends on agents failing to distinguish a run of good luck from a favorable structural shift in their environment." It is indeed true that Minsky refers occasionally to "conventional" assumptions which prove to be false, and to psychological phenomena such as "success breeds daring" (1975, p. 128). It is far from clear, however, that such features of the theory are in any way essential to the argument. The possibility that an optimizing approach to expectation formation may not be inconsistent with the theory arises because Minsky is dealing explicitly with what has come to be called a self-referential system (Marcet and Sargent, 1989), in which the actual law of motion describing the evolution of economic magnitudes is not independent of the perceived law of motion which guides the actions of agents. In such environments, the convergence to rational expectations trajectories even of extremely sophisticated learning procedures cannot be taken for granted. It is true that the FIH does not appear to be compatible with the RE hypothesis, in that firms persist in adopting liability structures which give rise to outcomes which violate the assumptions on the basis of which the liability structures were chosen. This occurs both in the case of excessive caution following a period of instability, and excessive boldness following a long expansion. Nevertheless, simply because a Minsky economy is not characterized by a movement along a rational expectations equilibrium trajectory, this does not force one to conclude that the agents described therein are in any meaningful sense irrational; the question must be raised within an explicit model of learning.
In retrospect, my attempt to answer this question fell woefully short. But the question itself seems worth addressing, as part of an attempt to explore the logical consistency of the financial instability hypothesis and clarify the assumptions on which it is based.


  1. Some thoughts on how to model FIH with rational expectations:

    Lenders know there is a cycle, but they don't know precisely where they are in the cycle. This generates a distribution of lenders, some more risk averse than others.

    There are some agents who exploit asymmetric information and seek to borrow money from the least risk averse lenders. These agents intend to make short term gains and declare bankruptcy.

    This may generate an equilibrium where the least risk averse lenders always go bankrupt -- and the only irrationality is that the lenders don't know where they are in the distribution of beliefs about the cycle. (i.e. the definition of a "belief" about the cycle is that everybody thinks his belief is average.)

    As long as there aren't too many predatory borrowers in the economy, it should be possible for lenders to expect to earn normal profits.

  2. Dear Professor Rajiv,

    I'd like to thank you for this post. I have linked to it here.

    I'm a young student of complex systems, economics, decision making, finance, and investing and a devout follower of your blog.

    Given our similar interests. If you have 5 minutes I would like to get your opinion of my blog.

    Thank you for your time I wish you the best of the holiday season.

    Miguel Barbosa

    Founder of

  3. "But is it really true that Minsky's theory depends on an assumption of significant and persistent departures from economic rationality on the part of firms?"
    The dependency is irrelevant since the assumption is valid. The irrationality of firms can be proven via the principle-agent problem:

    The incentive structure of a firm as a whole is very close to the maximization of expected profit. This does not match the incentive structure of the individual traders or money managers employed at the firm. The worst that can happen to an individual trader is that he is fired - his risk is capped. As long as excessive profits contribute to his bonus, he is strongly incentivized to make excessively risky bets: even bets that have negative expected profit! If his bets turn out well, he collects a sizable bonus. If not, well, it's not his dime.

    This fits in well with Minsky's theory of stability breeding instability. The mechanism of creeping appetites for risk is not the firms' defectively short memory, but their employees' correct assessment of how they are paid. It is noteworthy that in firms where risk is monitored (self-enforced or by outside auditors), the traders have every reason to sweep risk under the rug. One wonders how closely risk-appetites match the discovery of new accounting tricks.