Deflation, the Business Cycle and Depression: Is There a Link?

In a recent post on the blog of Cynicus Economicus, there is an interesting discussion on deflation:

http://cynicuseconomicus.blogspot.com/2009/08/deflation-scare.html

Deflation is a decrease in average prices throughout an economy. It is often accompanied by cuts in nominal wages. It is very important to emphasise this definition of deflation: by using the word “deflation” in this article, I am not talking about a decline in the prices of only one or two goods in an economy. Nobody, for instance, denies that specific price deflation in computers (caused by industry producing increasingly cheaper and better computers) is a good thing. But, when average prices across an economy fall in a significant way, you have general deflation, and this is a very different phenomenon from decreases in the prices of a few consumer goods that do not drag down the average price level.

With the exception of Japan in the 1990s, sustained and general price deflation is not something we in the West have seen since the 1930s.

The depressions of 1921–1922 and 1929–1933 were accompanied by severe deflation, and so were depressions in the classical gold standard era.

After World War II, deflationary busts were no longer a feature of the business cycle. The reason this happened was that the business cycles of the post World War II era have been fundamentally different from those before the 1930s. In essence, the gold standard era had its own distinctive business cycle. Before 1931/1933, the West was on gold standards of various types. The UK, for instance, was on the gold standard between 1819–1914 and a gold exchange standard from 1925–1931. The US had a de facto gold standard from 1834.

The age of the classical gold standard, when many other countries adopted the system, was roughly from the 1870s/1880s until 1914.

The gold standard era had its own particular type of business cycle. In that era, contractions in the economy (busts) were usually shortly preceded by the collapse of speculative bubbles and accompanied by general deflation. That is to say, as output collapsed and the economy contracted, deflation was a concomitant feature of the contraction.

I. Does Deflation Cause Depression?
It is certainly the case the price deflation was not in general the cause of gold standard depressions. There are few economists who seriously argue that general price deflation is a single, overarching and actual cause of contractions in the business cycle. Rather, price deflation was a regular characteristic (or symptom) of a contraction in the business cycle. As an economy contracted, at the same time it experienced deflation. The actual causes of recessions and depressions are varied and different from deflation, although deflation could in theory make a depression worse.

Since depressions and recessions in the gold standard era were nearly always deflationary, this is where the link between deflation and depression originates.

But the question whether sustained deflation during a contraction in the business cycle exacerbates depressions (rather than causes them) is, of course, a completely different question, and the volume of evidence suggests that deflation does indeed make depressions or recessions worse.

II. Has Deflation Accompanied a Growth in Output?
Yes, without a doubt. In the 19th century, there was a period of sustained deflation that lasted from 1873 to 1896. The period, however, was not one of continuous economic contraction: there were internal periods of economic growth and contraction (expansions and depressions). The entire period from 1873–1896 was not a “depression” in the accepted sense, but a period where prices showed a general trend towards deflation. The conventional explanation for this prolonged price deflation is that in this period there was an inadequate expansion in the money supply: money demand outstripped supply (money being tied to gold in this era).

To study this period, we can take England as an example. From 1873–1896, England experienced price deflation, but had a number of business cycles you can see here:

1873–1879, depression
1879–1883, expansion
1883–1886, depression
1886–1890, expansion
1890–1894, depression.

Thus there were three economic contractions (genuine downturns in the business cycle, with slumping production, unemployment etc) in this period, but there were two periods of expansion: during these booms output increased despite general deflation, and overall the period saw increased output. So it must be admitted that the idea that deflation only occurs during depressions or recessions is a myth.

Cynicus Economicus argues that

It is worth noting that, if wages were to remain static in monetary unit terms during a period of steady deflation, the recipient of the wages would find the purchasing power of their wage increasing.

This is correct. In the period from 1873–1896, nominal wages did not fall significantly or as fast as prices. This meant that real wages actually rose and living standards rose as well.

However, it should be noted that, apart from the period 1873–1896, expansions in the gold standard era tended to be inflationary. For example, the sustained boom that began in 1898 in most countries and that continued until 1913 was inflationary.

But Cynicus Economicus goes on to argue that

if wages were to remain static in monetary unit terms during a period of steady deflation, … the person would, in real terms, see an increase in their wealth. Even if the person’s wage were to decrease in a period of deflation provided that the decrease is less than the rate of deflation, they would still be seeing an increase in their wealth. Why such an outcome might be viewed as problematic is entirely unclear.

The fatal flaw in this argument is that it fails to take account of the effects of large amounts of debt (or other fixed contracts like leases) during unexpected deflationary periods. If nominal wages remain constant but prices of goods fall (and hence sales earnings), eventually this will cause profits to fall if companies have debt to service or rent to pay.

III. Debt Deflation: It Creates Deeper and Longer Recessions
Cynicus Economicus argues that

That depression might create deflation is not to say that the deflation is itself problematic. When looking at the deflation scare, it is a genuine puzzle that the scare has been allowed to gain so much traction. There is simply no evidence that a deflation would take the economy deeper into depression.

It is correct that deflation is not generally the initial cause of recessions. It could even be said that, under certain circumstances, deflation itself is not problematic.

But under a combination of certain factors, deflation will be catastrophic.

When there is a depression or recession, and deflation occurs, there are reasons why deflation can cause deeper economic contractions. The crucial point is that, if there is a very high level of debt before a recession begins, then deflation can have devastating effects.

Quite simply, Cynicus Economicus does not address the issue of profit and wage deflation, loss of consumer income, and the effects of continuing deflation.

It is the interaction of factors caused by deflation in a recession that can lead to a self-reinforcing downward spiral of prices, profits and wages.

The crucial factor is that the deflation continues. If prices fall, eventually profits fall as well, and employers must cut wages or reduce employment.
Because of wage “stickiness,” businesses will often be forced to reduce employment, rather than reduce wages. Debtors will suffer when they become unemployed and have no income.

Recessions can cause deflationary pressures. When demand falls and consumption falls sharply, first inflation falls through distress selling. If demand and consumption do not recover (or indeed become worse), this cost cutting caused by businesses reducing excess inventory will result in actual deflation. During this process unemployment rises and there will be downward pressure on wages. If there are steep cuts in wages, then incomes are reduced: this is the real cause of debt deflation: unemployment and cuts to wages.

There is both empirical and theoretical evidence that large amounts of debt in an environment of unanticipated wage and price deflation has disastrous effects on economic activity (Zarnowitz 1992: 156; Caskey and Fazzari 1987).

The economist Hyman Minsky (1892; 1986) has studied in detail how such debt deflationary spirals occur in the context of financial crises.

Debt contracts are set in nominal terms. As debtors see their incomes falling (or are laid off), some will eventually be unable to service debt, not only because their nominal wages have fallen, but also because they pay back debt with money of greater value.

Thus debt defaults and bankruptcies increase, money supply contracts, and there are further falls in demand. Debt deflation is a problem for businesses as well. As profits decline, business themselves are subject to much the same problem as individuals. Their debts become a much greater burden. This causes additional falls in output as business go bankrupt. The economy enters a vicious cycle.

The classic example of a deflationary spiral was 1929–1933. Moreover, if a government lowers interest rates in an attempt to stimulate the economy but cannot reduce it any further (because it will approach zero or become negative), then there is also a danger of entering a liquidity trap, if banks do not lend money, and hold excess reserves which they refuse to lend for investment.

This is the recipe for a Great Depression.

IV. Wage Stickiness: An Old Problem
According to Classical economics, downward wage flexibility was supposed to prevent an economy from ever falling into a deflationary spiral and collapse. But nominal wages, like loans or leases, can be fixed by contracts that do not take account of future deflation. As is well known, wage levels in practice tend to respond slowly to economic shocks, and they simply cannot be adjusted instantaneously: wages are “sticky” (Arnold 2008: 167).

But when deflation occurs and nominal wages are not adjusted downward, then profit margins fall (Kumar et al. 2003: 8). Businesses are faced with lower profits. In the face of wage stickiness, the first response tends to be job losses rather than wage cuts. But higher unemployment simply causes further collapses in demand. Once again a vicious circle has developed.

The problem of wage stickiness is a well known one in modern economics. People in general object to having their nominal wages cut. Even managers often dislike across-the-board pay cuts. Recent studies suggest that employers avoid pay cuts because they diminish workers’ morale, and then falling morale reduces productivity (Bewley 1999).

Wage stickiness, then, is a problem that happens even in free market economies. The economic theory that drastic cuts in wages will be able to cure depressions quickly (a feature of neoclassical economics and the Austrian school) is simply a fantasy that takes no account of the empirical evidence from the real world.

One extraordinary example of deflation making a depression worse was the depression of 1873–1879. In this time, in industrial economies like Germany, deflation was accompanied by falls in profits and cuts to nominal wages in industry that reduced real wages and living standards (Kitchen 1978: 159). This was the longest economic depression in recent history: it consisted of 65 months of economic contraction with deflation (Glasner and Cooley 1997: 148, 734).

V. The Solution to Wage Stickiness: Fiat Money!
Since people find it difficult to accept cuts in nominal wages, even if the real wages remain the same or actually rise during deflation, a practical solution is the use of fiat money: the central bank fights deflation with an expansion of the money supply.

If fiscal policy is used to inject the new fiat money into the economy, this will cause inflation and end the vicious circle. Putting people back to work raises output and stimulates demand. Employers find that they no longer have to face the problem of wage cuts as real wages will adjust through the value of money slightly decreasing through later inflation.

Thus fiat money and reflation (and of course fiscal policy) are pragmatic solutions to deflationary spirals and depressions.

VI. Money Does not Rise in Value During Disinflation (High to Low Inflation)
Cynicus Economicus argues that

A good example of this can be seen in private mortgages on housing. If a loan is taken out in a high inflation environment, the interest rate will be relatively high. The targeted central bank interest rate will be high, and the lenders will seek to account for the high inflation by charging a rate of interest that will overcome the devaluation of the money that they are lending, such that they can achieve a positive return. If the interest rate is fixed over a period of, for example, five years and at year four the rate of inflation has fallen by a half, the holder of the debt is effectively seeing the value of their debt inflating. The earlier rate of inflation was eroding the value of their overall debt, and this was accounted for in the interest rate. However, with inflation falling, their debt value is no longer declining at the same high rate, but they are still servicing the debt as if this were the case. Their payments in relation to the actual value of the debt have increased.

The specific effect of debt-deflation Cynicus Economicus is talking about here is when money rises in value through deflation. That is, if you pay back loans in money of higher value later (when it can purchase more), you are experiencing a specific aspect of debt deflation.

But, in the example Cynicus Economicus gives, the real issue is that the debtor is paying a higher real interest rate.

You can calculate the effect of higher real interest rates in this example:

In 2000, a business takes out a loan for five years at a 15% interest rate when inflation is 10% and the bank thinks it will stay at around 10% for some years. The real interest rate in 2000 is 5%. But inflation falls to 5% by 2003. The real interest rate has risen to 10%.

NIR = nominal interest rate
I = inflation rate
RIR = real interest rate.

Year NIR I RIR
2000 15% 10% 5%
2001 15% 10% 5%
2002 15% 9% 6%
2003 15% 5% 10%
2004 15% 5% 10%

But this effect is different from paying the money back when it is of greater value through deflation.

We can quote this explanation from Wikipedia:

Deflation is a sustained decrease in the general price level resulting in a sustained increase in the real value of money and other monetary items. Money and other monetary items are worth more all the time during deflation as opposed to being worth less all the time during inflation. Deflation is negative inflation. Disinflation is lower inflation. Prices are still rising during disinflation, but at a lower rate. The general price level still rises, but at a slower rate resulting in a continued, but lower rate of real value destruction in money and other monetary items. A lowering of inflation is not deflation but disinflation. Deflation means the general price level is not increasing at all, but, actually decreasing continuously and the internal functional currency – money - and other monetary items are worth more all the time. Deflation causes an increase in the real value of money and other monetary items. Inflation destroys real value in money. Disinflation destroys real value in money more slowly. Deflation creates real value in money.

http://en.wikipedia.org/wiki/Disinflation

The debt deflation effect I have talked about earlier is when you pay back your loan in money of greater value owing to deflation. (But of course paying higher real interest rates is also a part of the problem under deflation and, to this extent, the two situations are similar.)

But, even under disinflation (the move from higher inflation to lower inflation), the value of money is still falling, because it is only the rate of inflation that has changed. You are paying back your loan at a higher real interest rate, and are not subject to the specific debt deflationary effect mentioned earlier. That effect requires that the value of money has fallen through actual deflation.

VII. Does a Move from High Inflation to Low Inflation Cause Debt Deflation and Depression?
Of course not. In a booming economy, when there is a fall in the rate of inflation, this will not have the same effects as long-term, severe deflation in a recession where there is a large amount of debt.

A change in the inflation rate from 3.5% to 2% over one year and then stable inflation at 2% in a booming economy with low debt and high employment will not be a serious problem.

Disinflation is not actual deflation. Average prices are not falling.

The crucial factors that would cause a serious debt deflation and exacerbate a recession would be as follows:

(1) High debt levels before deflation
(2) A recession (for example, caused in part by the collapse of a bubble)
(3) Significant falls in demand and distress selling
(4) Nominal wages at too high a level
(5) Long-term, unexpected and severe deflation in goods produced in an economy
(6) Significant falls in profits, difficulty in servicing debt
(7) High unemployment, business failures, cuts to wages
(8) Further collapses in demand
(9) Then back to (6), (7) etc above.

This is a deflationary spiral. This is similar to what happened in the Great Depression.

A fall from high inflation to low inflation and then stable inflation in an essentially healthy economy does not have the same effect.


BIBLIOGRAPHY

Arnold, R. A., 2008. Economics (9th edn), Cengage Learning.

Atkeson, A. and P. J. Kehoe. 2004. “Deflation and Depression: Is There an Empirical Link?” American Economic Review 94.2 (Papers and Proceedings of the One Hundred Sixteenth Annual Meeting of the American Economic Association San Diego, CA, January 3–5, 2004). 99–103.

Bewley, T. F. 1999. Why Wages Don’t Fall During a Recession, Harvard University Press, Cambridge, MA.

Bordo, M. D., Erceg, C. J. and C. L. Evans. 2000. “Sticky Wages, and the Great Depression,” American Economic Review 90.5: 1447–1463.

Bordo, M. D. and A. Redish. 2004. “Is Deflation Depressing? Evidence from the Classical Gold Standard,” in R. C. K. Burdekin and P. L. Siklos (eds), Deflation: Current and Historical Perspectives, Cambridge, Cambridge University Press.

Caskey, J. and S. Fazzari, 1987. “Aggregate Demand Contractions with Nominal Debt Commitments,” Economic Inquiry 25: 583–597.

Eichengreen, B. J. 2002. Golden Fetters: The Gold Standard and the Great Depression, 1919-1939, Oxford University Press, New York.

Farrell, C. 2004. Deflation: What Happens When Prices Fall, HarperBusiness, New York.

Glasner, D. and T. F. Cooley (eds). 1997. Business Cycles and Depressions: An Encyclopedia, Garland Pub., New York.

Kitchen, M. 1978. The Political Economy of Germany, 1815–1914, Croom Helm, London.

Kumar, M. S. et al. 2003. Deflation: Determinants, Risks, and Policy Options, International Monetary Fund, Washington, D.C.

Minsky, H.P. 1982. Can “It” Happen Again? M.E. Sharpe, Armonk, NY.

Minsky, H.P. 1986. Stabilizing an Unstable Economy, Yale University Press, New Haven.

Smith, G. W. 2006. “The Spectre of Deflation: a Review of the Empirical Evidence,” Canadian Journal of Economics 39.4: 1041–1072.

Zarnowitz, V. 1992. Business Cycles: Theory, History, Indicators, and Forecasting, University of Chicago Press, Chicago.