Can It Happen Again?

Essays on Instability and Finance

Hyman Minsky

• Publisher: Routledge
• Year: 2016
• ISBN:
  9781317232490
• Edition: 1
• Topic: Negocios y empresa
• Subtopic: Negocios en general

In the winter of 1933, the American financial and economic system collapsed. Since then economists, policy makers and financial analysts throughout the world have been haunted by the question of whether "It" can happen again. In 2008 "It" very nearly happened again as banks and mortgage lenders in the USA and beyond collapsed. The disaster sent economists, bankers and policy makers back to the ideas of Hyman Minsky – whose celebrated 'Financial Instability Hypothesis' is widely regarded as predicting the crash of 2008 – and led Wall Street and beyond as to dub it as the 'Minsky Moment'.

In this book Minsky presents some of his most important economic theories. He defines "It", determines whether or not "It" can happen again, and attempts to understand why, at the time of writing in the early 1980s, "It" had not happened again. He deals with microeconomic theory, the evolution of monetary institutions, and Federal Reserve policy. Minsky argues that any economic theory which separates what economists call the 'real' economy from the financial system is bound to fail. Whilst the processes that cause financial instability are an inescapable part of the capitalist economy, Minsky also argues that financial instability need not lead to a great depression.

This Routledge Classics edition includes a new foreword by Jan Toporowski.

1

CAN “IT” HAPPEN AGAIN?

In the winter of 1933 the financial system of the United States collapsed. This implosion was an end result of a cumulative deflationary process whose beginning can be conveniently identified as the stock-market crash of late 1929. This deflationary process took the form of large-scale defaults on contracts by both financial and nonfinancial units, as well as sharply falling income and prices.¹ In the spring of 1962 a sharp decline in the stock market took place. This brought forth reassuring comments by public and private officials that recalled the initial reaction to the 1929 stock-market crash, as well as expressions of concern that a new debt-deflation process was being triggered. The 1962 event did not trigger a deflationary process like that set off in 1929. It is meaningful to inquire whether this difference is the result of essential changes in the institutional or behavioral characteristics of the economy, so that a debt-deflation process leading to a financial collapse cannot now occur, or merely of differences in magnitudes within a financial and economic structure that in its essential attributes has not changed. That is, is the economy truly more stable or is it just that the initial conditions (i.e., the state of the economy at the time stock prices fell) were substantially different in 1929 and 1962?

I. GENERAL CONSIDERATIONS

The Council of Economic Advisers’ view on this issue was stated when they remarked, while discussing fiscal policy in the 1930s, that “… whatever constructive impact fiscal policy may have had was largely offset by restrictive monetary policy and by institutional failures—failures that could never again occur because of fundamental changes made during and since the 1930s.”² The Council does not specify the institutional changes that now make it impossible for instability to develop and lead to widespread debt-deflation. We can conjecture that this lack of precision is due to the absence of a generally accepted view of the links between income and the behavior and characteristics of the financial system.

A comprehensive examination of the issues involved in the general problem of the interrelation between the financial and real aspects of an enterprise economy cannot be undertaken within the confines of a short paper.³ This is especially true as debt-deflations occur only at long intervals of time. Between debt-deflations financial institutions and usages evolve so that, certainly in their details, each debt-deflation is a unique event. Nevertheless it is necessary and desirable to inquire whether there are essential financial attributes of the system which are basically invariant over time and which tend to breed conditions which increase the likelihood of a debt-deflation.

In this paper I will not attempt to review the changes in financial institutions and practices since 1929. It is my view that the institutional changes which took place as a reaction to the Great Depression and which are relevant to the problem at hand spelled out the permitted set of activities as well as the fiduciary responsibilities of various financial institutions and made the lender of last resort functions of the financial authorities more precise. As the institutions were reformed at a time when the lack of effectiveness and perhaps even the perverse behavior of the Federal Reserve System during the great downswing was obvious, the changes created special institutions, such as the various deposit and mortgage insurance schemes, which both made some of the initial lender of last resort functions automatic and removed their administration from the Federal Reserve System. There should be some concern that the present decentralization of essential central bank responsibilities and functions is not an efficient way of organizing the financial control and protection functions; especially since an effective defense against an emerging financial crisis may require coordination and consistency among the various units with lender of last resort functions.

The view that will be supported in this paper is that the essential characteristics of financial processes and the changes in relative magnitudes during a sustained expansion (a period of full-employment growth interrupted only by mild recessions) have not changed. It will be argued that the initial conditions in 1962 were different from those of 1929 because the processes which transform a stable into an unstable system had not been carried as far by 1962 as by 1929. In addition it will be pointed out that the large increase in the relative size of the federal government has changed the financial characteristics of the system so that the development of financial instability will set off compensating stabilizing financial changes. That is, the federal government not only stabilizes income but the associated increase in the federal debt, by forcing changes in the mix of financial instruments owned by the public, makes the financial system more stable. In addition, even though the built-in stabilizers cannot by themselves return the system to full employment, the change in the composition of household and business portfolios that takes place tends to increase private consumption and investment to levels compatible with full employment.

In the next section of this paper I will sketch a model of how the conditions compatible with a debt-deflation process are generated. I will then present some observations on financial variables and note how these affect the response of the economy to initiating changes. In the last section I will note what effect the increase in the relative size of the federal government since the 1920s has had upon these relations.

II. A SKETCH OF A MODEL

Within a closed economy, for any period

which can be written as:

where S – I is the gross surplus of the private sectors (which for convenience includes the state and local government sector) and T – G is the gross surplus of the federal government. The surplus of each sector ζ_j(j = 1 … n) is defined as the difference between its gross cash receipts minus its spending on consumption and gross real investment, including inventory accumulations. We therefore have

Equation 3 is an ex post accounting identity. However, each ζ_j is the result of the observed investing and saving behavior of the various sectors, and can be interpreted as the result of market processes by which not necessarily consistent sectoral ex ante saving and investment plans are reconciled. If income is to grow, the financial markets, where the various plans to save and invest are reconciled, must generate an aggregate demand that, aside from brief intervals, is ever rising. For real aggregate demand to be increasing, given that commodity and factor prices do not fall readily in the absence of substantial excess supply, it is necessary that current spending plans, summed over all sectors, be greater than current received income and that some market technique exist by which aggregate spending in excess of aggregate anticipated income can be financed. It follows that over a period during which economic growth takes place, at least some sectors finance a part of their spending by emitting debt or selling assets.⁴

For such planned deficits to succeed in raising income it is necessary that the market processes which enable these plans to be carried out do not result in offsetting reductions in the spending plans of other units. Even though the ex post result will be that some sectors have larger surpluses than anticipated, on the whole these larger surpluses must be a result of the rise in sectoral income rather than a reduction of spending below the amount planned. For this to take place, it is necessary for some of the spending to be financed either by portfolio changes which draw money from idle balances into active circulation (that is, by an increase in velocity) or by the creation of new money.⁵

In an enterprise economy the saving and investment process leaves two residuals: a change in the stock of capital and a change in the stock of financial assets and liabilities. Just as an increase in the capital-income ratio may tend to decrease the demand for additional capital goods, an increase in the ratio of financial liabilities to income (especially of debts to income) may tend to decrease the willingness and the ability of the unit (or sector) to finance additional spending by emitting debt.

A rise in an income-producing unit’s debt-income ratio decreases the percentage decline in income which will make it difficult, if not impossible, for the unit to meet the payment commitments stated on its debt from its normal sources, which depend upon the unit’s income. If payment commitments cannot be met from the normal sources, then a unit is forced either to borrow or to sell assets. Both borrowing on unfavorable terms and the forced sale of assets usually result in a capital loss for the affected unit.⁶ However, for any unit, capital losses and gains are not symmetrical: there is a ceiling to the capital losses a unit can take and still fulfill its commitments. Any loss beyond this limit is passed on to its creditors by way of default or refinancing of the contracts. Such induced capital losses result in a further contraction of consumption and investment beyond that due to the initiating decline in income. This can result in a recursive debt-deflation process.⁷

For every debt-income ratio of the various sectors we can postulate the existence of a maximum decline in income which, even if it is most unfavorably distributed among the units, cannot result in a cumulative deflationary process, as well as a minimum decline in income which, even if it is most favorably distributed among the units, must lead to a cumulative deflationary process. The maximum income decline which cannot is smaller than the minimum income decline which must lead to a cumulative deflationary process, and the probability that a cumulative deflationary process will take place is a nondecreasing function of the size of the decline in income between these limits. For a given set of debt-income ratios, these boundary debt-income ratios are determined by the relative size of the economy’s ultimate liquidity (those assets with fixed contract value and no default risk) and the net worth of private units relative to debt and income as well as the way in which financial factors enter into the decision relations that determine aggregate demand.

If the financial changes that accompany a growth process tend to increase debt-income ratios of the private sectors or to decrease the relative stock of ultimate liquidity, then the probability that a given percentage decline in income will set off a debt-deflation increases as growth takes place. In addition, if, with a given set of debt-income ratios, the net worth of units is decreased by capital or operating losses, then both the maximum decline in income which cannot and the minimum decline in income which must generate a debt-deflation process will decrease. If the economy generates short-term declines in income and decreases in asset values in a fairly routine, regular manner then, given the evolutionary changes in financial ratios, it is possible for an initiating decline in income or a capital loss, of a size that has occurred in the past without triggering a severe reaction, to set off a debt-deflation process.

A two sector (household, business) diagram may illustrate the argument. Assume that with a given amount of default-free assets and net worth of households, a decline in income of ΔY₁ takes place. For ΔY₁ there is a set of debt-income ratios for the two sectors that trace out the maximum debt-income ratios that cannot generate a debt-deflation process. There is another set of larger debt-income ratios which trace out the minimum debt-income ratios which must generate a debt-deflation process when income declines by ΔY₁. For every debt-income ratio between these limits the probability that a debt deflation will be set off by a decline in income of ΔY₁ is an increasing function of the debt-income ratio.

The isoquants as illustrated in Figure 1 divide all debt-income ratios into three sets. Below the curve A-A are those debt-income ratios for which a decline in income of ΔY₁ cannot lead to a debt deflation. Above the l...