Category Archives: Capital Theory

Unemployment’s Demand Side

About four months ago, I wrote on a for-fun empirical model, to test whether cyclical unemployment is predominately supply- or demand-driven. At the time, the results suggested that aggregate demand was the main determinant, although “labor freedom” (the only statistically-significant supply-side factor) could explain quite a bit in countries where unemployment is highest. I wanted to come back and see what the results look like if I try to incorporate the Phillips curve. Also, it would be interesting to know whether a demand shortage’s effect accumulates as “labor freedom” decreases.

The model is,

Unemployment Rate = β0 + β1Labor Freedom + β2Business Freedom + β3Fiscal Freedom + β4Property Rights + β5%ΔMoney + β6%ΔMoney2 + β7%ΔMoney×Labor Freedom + ε

Unemployment data is from the IMF, money growth data is from the World Bank, and the Heritage Foundation provides the rest. Definitions and rationales for the variables,

(a) Labor freedom: A proxy for labor market flexibility. Where labor regulations increase hiring costs, we expect greater unemployment.

(b) Business freedom & property rights: These were included to test the regime uncertainty thesis. According to the theory, lower business freedom and ambiguous property rights are disincentives to investment (and indirectly, employment).

(c) Fiscal freedom: This is a measure of tax burden. One belief is that higher taxes are a disincentive to invest. There is also a related argument that reducing taxes is stimulating, because it increases private spending. Either way, the supply-side argument expects positive correlation between tax burden and unemployment.

(d) Percent change in the money supply: This is the aggregate demand variable — and the data i for base money. I agree, maybe NGDP may have been a better choice. I’m not entirely convinced, because I think changes in money supply might test aggregate demand as a policy more directly. There is probably a choice that’s better than both of them, because changes in aggregate income can be caused by both supply- and demand-side factors.

Unemployment v. Money Growth (version 2)

The Phillips curve is introduced through the quadratic. The accumulation effect is represented by the interaction term %ΔMoney×Labor Freedom. Surprisingly, the interaction term is statistically insignificant. At least, it was surprising to me, because, for countries with very high unemployment rates (e.g. Spain and Greece), the coefficients for the linear approximation did not seem able to explain, on their own, the extent of their unemployment. So, when these two factors interact, maybe their effect is much worse than simply their sum. The data says no.

The two statistically significant variables are the quadratic term and labor freedom.

Unemployment Rate = 23.05 + –0.1021Labor Freedom + –.5726%ΔMoney + 0.0138%ΔMoney2

Introducing a Phillips curve actually very slightly decreases the effect of monetary policy. The prediction is that a one percentage point increase in rate of growth of the money supply will reduce the unemployment rate by –.5576 + .0276(%ΔMoney) percentage points. What this implies is that there are diminishing returns to monetary policy, and that as monetary policy growth rates become smaller, the negative effect becomes greater. The data is in line with the position that the worst monetary policy — whether public (central bank) or private (e.g. a clearinghouse issuing its own emergency notes) — is pro-cyclical, but that excess supplies of money at first have only a relatively minor negative effect on unemployment and then a gradually growing positive effect. Here, negative and positive have reverse meanings, because a positive coefficient implies a growing unemployment rate.

The marginal effect of labor freedom has also slightly increased, from –0.0961 to –0.1021. The original test, therefore, slightly underrepresented the role of inflexible labor markets. Still, labor inflexibility explains relatively little. For example, 2012 U.S. had a ranking of 95.5 out of 100; the most it could increase its labor flexibility is by 4.5 points, or a decrease in the unemployment rate by –.4595 percentage points. For cases like these, the coefficient is probably overstated, because it’s probably not the case that an increase in labor freedom from 95 to 96 is as significant as a change from 50 to 51. Likewise, if Greek labor markets were as flexible as U.S.’, the 2012 Greek unemployment rate would be 5.4521 percentage points lower. Even in cases where labor inflexibility is significant, it cannot explain even the majority of cyclical unemployment.

Because it’s interesting to some, that neither business freedom or property rights are statistically significant suggests that regime uncertainty is not a relevant explanation for cyclical unemployment — at least, not in this recession.

Piketty’s Meat and Bones

I have yet to finish reading Capital in the Twenty-First Century. There are many reviews already out, many of them very much worth reading. Two critical reviews by Ryan Decker and Josh Hendrickson. There is also Paul Krugman’s, probably the most widely discussed review to date; Krugman makes some good points against Piketty, but also covers some of the book’s strengths.

The heuristic the book sells to the reader is r > g. That is, if the return on capital is greater than the rate of economic growth, inequality will increase. The concept, explained like that, is somewhat cryptic. It’s an easy heuristic for those who don’t want to get bogged down in the theoretical argument. The theoretical argument is actually not that complicated (theory makes up a very, very small minority of the book). Piketty makes it near the end of the sixth chapter of his book. His argument is that if the elasticity of substitution between capital and labor is > 1, the share of income accruing to capital will grow relative to that of labor.

You actually don’t have to buy the book to access the theory. In my opinion, the theory is not well communicated in the book. It’s available in the free technical appendix. Here is that appendix for chapter six (in case it doesn’t take you directly, the meat and bones starts on page 37),

Download the PDF file .

ABCT and the Price Level

I remember my monetary theory professor using an aggregate demand/supply graph to illustrate supply-side theories of the business cycle, where a supply shock shifts the supply curve to the left. All else equal, a negative supply shock will lead to an increase in the price level — the same amount of money chasing less goods. This is how some people interpret Austrian business cycle theory (ABCT): there is a negative supply-side shock to the capital stock. It follows that the price level for capital goods rises. But, the evidence suggests the exact opposite, falling prices in the capital (intermediate) goods sector. That interpretation is wrong. ABCT is a demand-side theory, and it predicts just what the evidence shows, a falling price level for capital goods.

I also discuss what predictions ABCT makes regarding the general price level. The simple answer: none. This is why the ABCT cannot explain the depth of a recession on its own. I make my case in the second part of the post.

If you think that Hayek’s theory is about a supply-shock, read “The Maintenance of Capital.” The article presents a theory of the value of the heterogeneous capital stock. It’s related to Hayek’s business cycle research, in that it clearly frames the issue in the context of aggregate demand. Note his discussion of “capital consumption.” Capital consumption is not the same thing as a supply-side shock to the capital stock. Remember, the value of the capital stock is imputed (derived) from the final product. A rise in the price of an input is caused by an increase in demand for that input’s output. A fall in the demand for the final product, leads to a fall in the value of those inputs — capital consumption. This is important to keep in mind, because this is one of the basic parts to ABCT.

Let’s call an economy “neutral” when the allocation is optimal (don’t worry about frictions and imperfections). Picture an economy as a production possibilities frontier (PPF), with capital goods on one axis and consumer goods on the other. The PPF represents potential mixtures of output, given some given amount of resources. Not all points on the PPF are equally valuable. The neutral economy produces where the opportunity cost is lowest, which is a specific point on the PPF. Essentially, ABCT predicts that an excess supply of money will cause an economy to become non-neutral, by producing on some other point along (or inside) the PPF. This is shown in the graphic below, where the PPF is on the right — consumer goods on the y-axis and capital (or producer) goods on the x-axis. There are two marked places on the PPF; the uppermost circle is the original mix of output, and the lower circle is the new mix of output.

Output Combinations and the Hayekian Triangle

(Don’t worry about the Hayekian triangle on the left panel.)

Forget about central banks and interest rates, and focus on an excess supply of money. Hayek and Mises believed excess supplies of money to be non-neutral. Specifically, they thought excess supplies of money changed the prices of capital goods relative to those of consumer goods. The stylized process by which this happens is that new money is introduced through the loanable funds market, as banks extend loans. That borrowed money is used to buy certain inputs, raising the price of those inputs, and thus the profitability of producing them. The economy moves towards producing somewhere lower on our PPF curve, above (the second point on the curve). This means less consumer good output, since more resources are being used to manufacture intermediate (capital) goods. But, the level of consumption remains the same, so as the supply of consumers’ good falls, their price increases. This sets off a process of reversion, where the structure of production has to re-adapt to society’s time preference (intertemporal equilibrium).

Where is the shock to the capital stock? As amount of current inputs going into the production of future inputs increases, it changes the nature of capital. If the value of capital is imputed from the final product, and the final product capital is being used for changes, then the value of the capital stock will change. The excess supply of money, by raising the price of inputs, creates “false” profits. To keep it simple, “false” profits are non-neutral. These profits will make the new structure of production seem of high value. But, the reversion process corrects this, and there is a change in the pattern of demand. The investment during the boom is “malinvestment,” and the capital stock loses much of its value (it’s consumed), by virtue of the fact that the “false” demand for its output collapses.

Note that the prediction is a fall in the value of the capital stock. The physical stock of capital remains the same. The prices of capital goods, however, collapse. The type of inputs produced by the boom-time structure of production are not as useful as they first seemed. The range of productive uses narrows, and if the capital is highly specific it might become completely useless (Hayek makes this point explicitly in Prices and Production, and references the debate on idle resources).

To put the same point differently, imagine a supply/demand graph. The price of a capital good is decided at the margin at which the supply and demand curves meet. ABCT predicts that the demand curve will shift to the right during the boom, and then to the left during the process of reversion. Thus, the bust coincides with a falling price level for capital goods. The supply-side shock, just to be able to compare and contrast, predicts that the bust coincides with a leftward shift of the supply curve, or an increase in the price level for capital goods. The intuition behind that process is that as the amount of a type of capital decreases, the least valuable end that last input is useful towards increases in value (since there is less capital, there are less ends that can be satisfied — and, optimally, the most valued ends are satisfied first). This is not what ABCT predicts. Rather, ABCT says that, during the bust, the least valued end will decrease in value, because, as it turns out, what producers thought were highly valued ends were just distorted profit signals, caused by an excess supply of money.

The theory fits the empirical evidence, at least in this respect,

PPI Intermediate Materials 2001-2012

What about the general price level (consumers’ + producers’ goods)? ABCT is somewhat indeterminate in this regard, because capital consumption occurs while the price of consumers’ goods increases. It’s also ambiguous whether the excess supply of money will end through a rise in the price level, or through some kind of aggregate reflux mechanism. What brings about the primary effects of the ABCT are not changes in the money stock, but changes in relative prices. To explain a decline in the general price level, Austrians usually employ another theory, typically referred to as “secondary deflation.” Because this is no longer really within the scope of the ABCT, Austrians are somewhat vague on this point. But, I think this point is worth discussing, because it shows that ABCT most likely cannot explain a major business cycle on its own.

Rothbard provides an entire section (pp. 14–19) to explaining secondary deflation in America’s Great Depression. He employs a very strict quantity theorist definition of inflation/deflation: an increase/decrease in the money supply (although, I’m not sure this is consistent with some of the causes of deflation he discusses). Some of the causes he gives assume a gold standard, relevant then, but not now. Still, I think they can be generalized. He cites an outflow of gold (or, more broadly, let’s call it capital), forcing banks to contract their balance sheets. He also cites debt liquidation, where credit simply cannot be repaid. Finally, a third force is the increase in the demand for money. The deflationary process is, strictly speaking, separate from the process described by ABCT; related to it, but not directly. Rothbard considers the deflationary process beneficial. In the preface to “Monetary Theory and the Trade Cycle,” Hayek also distinguishes deflation from the process described by ABCT, but writes that “an indefinite continuation of that deflation would do inestimable harm.”

There are multiple margins to judge deflation on. Rothbard’s point, which is valid, is that what matters are relative prices, not general prices, so that deflation doesn’t necessarily cut into profits — especially if capital good prices fall faster than consumer good prices (which is empirically true). I will come back to this shortly. Other channels through which deflation may not be beneficial include those akin to Fisher’s theory of debt deflation. A New Keynesian model will also tell you that a non-inflationary environment forgoes the benefit of inflation to reduce the demand for money, if we are in a liquidity trap, where interest-bearing assets and money are equally attractive, because of low interest rates. I’m not going to comment on the validity of these theories. They are nevertheless worth considering, and I will say why that is below.

Debt deflation and liquidity traps aside, knowing why Rothbard is somewhat wrong is important. Note a claim he makes on the height of unemployment caused by an Austrian business cycle,

Since factors must shift from the higher to the lower orders of production, there is inevitable “frictional” unemployment in a depression, but it need not be greater than unemployment attending any other large shift in production. In practice, unemployment will be aggravated by the numerous bankruptcies, and the large errors revealed, but it still need only be temporary.

America’s Great Depression, p. 14.

This is, by the way, the exact claim Paul Krugman makes in his criticism of ABCT, what he calls the “hangover theory.” The initial boom period requires substantial capital readjustment, as well, but during this period the unemployment rate is relatively low — either around its “natural” level (~4–5 percent), or below. I think the critique is not as strong as some make it out to be, because the bust is a sudden change in relative demand, with significant capital consumption. During the boom, there is competing demand for consumer and capital goods, so one sector doesn’t necessarily suddenly become unprofitable. The shift in the allocation of labor is much less painful during the boom. Still, Krugman (and Rothbard) have a point: ABCT cannot explain the depth of unemployment very well.1 You need an additional story, and the scapegoat many Austrians like is government, bad supply-side policy and regime uncertainty. But, blaming government is not always right.2

Explaining the direction the general price level moves in is relevant to explaining the depth of unemployment. Putting debt deflation and liquidity traps aside, deflation is good if it’s neutral. If the demand for money increases and deflation is non-neutral, the unemployment rate and the output gap will grow. Assume the money stock at time t is M1, but the demand for money at time t+1 calls for a stock equal to M2. If the supply of money doesn’t increase, there will be a demand shortage. There will be people who want to increase their cash balances by selling non-money assets, including labor and goods/services. But, there isn’t enough money to meet their demand for it. If deflation is neutral, a falling price level will cause a falling demand for money. If deflation is not neutral, the demand for money remains the same there is a shortage of money, and therefore of demand, and trade that would have taken place does not — output falls. Deflation can be non-neutral if some key prices don’t fall to clear the market, including the price of labor and, perhaps, other capital goods that are heavily complimentary to labor. This is monetary (dis)equilibrium theory. You need this theory to explain the depth of the recession.3

Austrian business cycle theory does not offer us ideas on the direction the general price level takes during a bust. Empirically, deflation is the norm. ABCT cannot explain this deflation, except through use of alternative theories. Further, while the ABCT explains the boom and the direct causes of the bust, it cannot explain other elements of the bust that are caused by confounding factors. But, ABCT definitely does not predict an increase in the price level of capital goods, which is what would happen in a supply-side shock. Rather, it predicts declining capital good prices, which fits the evidence.



1. I discussed a similar criticism in an August 2011 article, “Rethinking Depression Economics.” I think that much of my insight is valuable, but, at the time, I had not considered the role of monetary equilibrium to the extent I should have.

2. Bad supply-side policy can explain a lot of unemployment. The high unemployment of the Great Depression is explained in large part by bad supply-side policy. Alternative theories should be seen as complimentary. Look at the difference in unemployment between Spain and the United States, both of which have similar output gaps. Spain is an example of bad supply-side policy. The U.S., however, is an example of bad demand-side policy. As the unemployment rate falls, the output gap remaining equal, supply-side theories of unemployment have less explanatory power.

3. Hayek understood this. Apart from his comments on the negative effects of deflation, see “The Gold Problem” (published in Good Money, Part I) and “Monetary Nationalism and Monetary Stability.” I think he realized the power of Hawtreyian (?) explanations of the Great Depression, which focused on international gold flows and reserve hoarding by part of central banks. He wanted to synthesize these theories with his own.

What Makes Austrian Business Cycle Theory Attractive

Tyler Cowen links to a recent paper which explores credit expansion and financial crashes. Cowen comments that Austrian Business Cycle Theory (ABCT) needs more Minsky. I disagree. It reminds me of a passage that I had highlighted in Piketty’s Capital in the Twenty-First Century, on bubbles and self-fulfilling expectations. One thing that makes ABCT so attractive to me — one of the most attractive facets of Austrian economics, in general — is the emphasis on rule-following behavior. This requires interpreting Hayek’s 1930s work on industrial fluctuations through a 1940s Hayekian lens. One way to interpret the business cycle is to ask why certain rules work for some period of time, which we can call the “boom,” and why these rules suddenly create erroneous decisions during another period of time, the “bust.”

Piketty offers a conventional psychological theory of the self-fulling bubble,

[T]hese anticipated future prices themselves depend on the general enthusiasm for a given type of asset, which can give rise to so-called self-fulling beliefs: as long as one can hope to sell an asset for more than one paid for it, it may be individually rational to pay a good deal more than the fundamental value of that asset (especially since the fundamental value is itself uncertain), thus giving in to the general enthusiasm for that type of asset, even though it may be excessive. That is why speculative bubbles in real estate and stocks have existed as long as capital itself; they are consubstantial with history.

— p. 172.

The above is representative of a fairly common way of explaining bubbles and financial crises. Other explanations, such as Minsky’s might be more sophisticated, but they essentially grasp towards the same idea. I find their insight to be somewhat trivial, because they are vague theories that simplify agents’ behavior in financial markets to “they buy what they think everyone else will buy.” But, and this is an empirical question, I doubt that very many investors make their choices based on the expectation that an asset will continue to rise in value, despite that price diverging from “fundamentals.” Instead, what we see is investors justifying their decisions based on their reading of the “fundamentals” — i.e. the national housing market will not suffer a bust —, only to find out that their interpretation is wrong.

What ABCT does, once fully synthesized with Hayek’s later work on knowledge, institutions, and rule-following behavior, is embrace rational, informed decision-making in financial markets, and then explain how these rules can lead to welfare reducing outcomes if these rules are “distorted.” Hayek’s early model simplified all of these behavior-constraining rules to profits. Investors chase profits, because profits signal healthy, strong investment. Where assets grow in value is where there are profits. To investors, these increased revenue streams are the “fundamentals.” They invest in an asset because that business, or that market, is doing well. Any theory of the business cycle has to explain why these expectations, and the “fundamentals” (the heuristics) that guide them, are suddenly reversed — why do markets, which usually read the “fundamentals” so well, fail in bulk?

Minsky needs ABCT, not the other way around. The ABCT is, so far, the most complete demand-side theory of reversed expectations. I will refrain from claiming it’s the best, because so far maybe it hasn’t been compared to adequate alternatives. And, undoubtedly, ABCT is far from perfect, itself. You also need to take an empirical approach, because the lessons of ABCT have to be generalized — the narrow model is not perfectly applicable (as is the case with every theory). But, the point I want to make is that the credit-boom aspect is not what’s unique to ABCT. You can find that in Minsky. Other economists have made similar connections (e.g. Alan B. Taylor). What’s unique to ABCT is how it puts the story together by trying to explain investors’ behavior, and why this behavior turns out to be systemically suboptimal. It provides a sophisticated explanation, and I don’t think Minsky — or others who rely on psychological, rather than institutional, explanations — succeeds to the same degree.

Spain’s Cyclical Disorder

I like to read the comments people leave in articles in Spanish sports papers. I find them funny. There’s typically a decent mix of inter-team hate, intra-team hate, and general conspiracy theory. One can find a similar mix in the press of other countries, but not nearly to the same extent as in Spain. What these comments reveal — although there may be some sample bias — is an emerging culture of mistrust, caused in large part by ongoing depression in Spain and political corruption. Strong economies are built on trust, and this emerging culture in Spain can undermine forces of recovery in a country already handicapped by bad structural and counter-cyclical policy.

In “Economics and Knowledge,” Hayek developed a theory of equilibrium incorporating the role of expectations. In a division-of-labor where each agent relies in large part on others, an equilibrium occurs when the expectations of these agents coincide. For example, you, generally speaking, would only continue to work if you were certain that you would receive a paycheck at the end of the week. Businesspeople have certain expectations of the demand for their product, and of profitability. In the real world, expectations are often wrong — businesses fail, stock markets crash, et cetera. In equilibrium expectations are in harmony, so that everyone’s plans coincide.

The term “coordination” is essentially synonymous with “equilibrium.” For disparate expectations to jive, plans must be coordinated. When plans don’t coincide, there’s discoordination. Coordination differs from equilibrium, however, in that it has a more ambiguous definition. It might be useful to think about it as a spectrum. Plans may not be perfectly coordinated, but people’s actions are not always total failures. In fact, most of the time, things work out. When you take your car to the mechanic, he’s there; your paycheck arrives on time; your landlord expects and gets his rent by the monthly deadline; et cetera. We can also think about an economy having a certain degree of order, where equilibrium is maximum order. The better coordinated plans are, the higher the degree of order.

The concept of order can be applied to business cycle theory. Expectations have played a central role in business cycle research since the 1930s. Most firms do not expect a sharp decline in demand. Those that do typically mistime their preparation. A clear example is the 2007–09 financial crisis. Investment banks’ plans did not work out when a significant fraction of their assets dropped in value. The small businesses which went bankrupt had their expectations reverse. In mid-2006, few people thought that they had a large probability of losing their jobs in 2008–09. The business cycle is a period of discoordination — a loss of order.

Spain’s recession, following a housing bubble, has been particularly bad. Indeed, it feels more like a depression. Spain’s unemployment rate is over 26 percent. Much like during the Great Depression, a large fraction of unemployment is the result of bad policy. The IMF estimates Spain’s average outgap gap for 2013 at –4.338, which is comparable to that of the United States. Yet, the U.S.’ unemployment rate is “only” at 6.7 percent. Bad supply-side policy, in conjunction with bad demand-side policy, has made a disaster of an already bad situation. Spain’s structural problems, of course, have existed for a long time, but they were partly hidden by the boom of the early and mid-2000s. Schneider (2011) estimates that the value of production in Spain’s informal (extralegal) sector has been greater than one fifth of Spain’s GDP since ~1994, falling to 19.3 percent of GDP in 2007 — but, no doubt larger after the crash (although, in part because of a decline in GDP). Large extralegal sectors are typically a sign of bad supply-side policy.

Problems in Iberia go beyond bad policy. Spain’s democracy is more fragile than some people realize. The country has passed a few tests for robustness — such as the (quasi-abortive) February 1981 coup —, and high growth during the late 1990s and for much of the 2000s seemed promising. Indeed, many in Spain were looking forward to the day that the country would replace Canada in the G8! But, the country’s economic woes have been joined with corruption issues in government — and the problem crosses party lines: bribery, illegal payments, money laundering, et cetera. Further, regional banks funded local programs, and the relationship between these banks’ balance sheets, their health, and the political support they receive is unclear. These scandals have made it hard for Spaniards to trust their governments, whether national or local. The range of this distrust has spread to the royal family — especially after legitimate corruption charges against some of its members — and elsewhere.

The extent of distrust can be seen in Spain’s changing football (soccer) culture. Referees are always criticized. It’s hard to find an article on a big Premier League game without reading one manager’s criticism of the referee’s handling of a game. There are constant complaints against refereeing in all sports. But, in Spain, it’s especially bad, because complaints have given way to conspiracy theories. There is this notion that the governing body in Spain, RFEF, implicitly or explicitly supports a status quo that favors Barcelona and Real Madrid, the two highest earning football teams in Spain. The theories go as far as to claim that referees make their calls to purposefully help the the two “big teams” (los grandes). And, of course, between the two big teams fans accuse the other club of corruption. Even players have made these accusations — Cristiano Ronaldo and Sergio Ramos, players for Real Madrid, made comments to this effect after last weekend’s match against Barcelona.

There is, in general, an emerging culture that sees corruption and scandal everywhere. Bonds of trust are disintegrating. This can have dire long-term consequences. Trust, in some sense, is an institution; or, it’s a value manifested in certain institutions. Businessmen spend a lot of time and money establishing a relationship of trust with their clients. They offer warranties, they make sure to always leave their customer satisfied (hoping for the relationship to be repeated over time), et cetera. When there is no trust, the economy suffers. The transaction costs to some exchanges can become too high, and trades which would have led to gains for all parties involved simply don’t take place. To put the role of trust into perspective, recall the 2011 paper by Nunn and Wantchekon, finding evidence that slavery created a culture of mistrust, holding development in sub-Saharan Africa back. This is happening in Spain. While political corruption is hard to compare to slavery, the emerging culture in Spain may have dire economic (and political) consequences for years to come.

Mistrust might make political reform more difficult. Note how untrustworthy Ukrainian protestors have been of the “reformed” Ukrainian government. The same is true of Spain, although in different areas and to a different degree. Governments with fragile relationships with their people are going to have a hard time passing structural reforms that may not produce immediate results. Mistrust also makes a situation ripe for regime uncertainty. Uncertainty over the political climate makes business more difficult and less attractive, especially if this business requires large investments. Mistrust can spill over into the private sector in other ways around. Large businesses — which tend to be villainized when it’s the worst off who suffer the most — can lose demand for their product, even if their product confers to the consumer the highest satisfaction relative to other choices. Mistrust can even hurt relationships between small businesses, and even between small time vendors and their customers.

The effects of the business cycle, especially if these effects are made worse by bad policy and political corruption, can influence an economy for many years thereafter. They can make recovery more difficult. When a people lose faith in their institutions of governance, this mistrust tends to spread to other sectors of life, including business. It itself is corrupting, because it changes the institutions which guide economic processes. It limits trade, and when trade is limited we lose out on gains from trade. Countries like Spain (and Greece) are undergoing this transformation, and if this process isn’t reversed, the consequences can become more dire than they already are.

Keynesian Stimulus, and Hayek

John Aziz writes,

But Keynes’ theory has nothing whatever to do with the size of the government. It is really about the timing of government spending. Keynes thought that against-the-grain fiscal policy could mitigate the business cycle, which he viewed as a natural outgrowth of the wild animal spirits of the market. When the economy is booming — unemployment low and growth strong — Keynes advised that government should reduce spending and run a surplus, thereby dampening demand and lowering the chance of the economy overheating. And when the economy is in recession, he advised that government should increase spending and run a large deficit, to boost demand and push the economy back toward growth.

This reminds me of two things. First, while not specifically on fiscal stimulus (which I don’t support), Aziz’ article reminded me of a blog post (of my own) with a very similar message, “What is Limited Government.” While I don’t put in these terms, I think this summarizes my idea pretty well. Just like we can only talk about excess or a shortage of demand in the context of some equilibrium, we can only talk about too much or too little government in the context of an equilibrium, or the “right amount” of government — and the latter equilibrium, like all equilibria, are not fixed.

Second, it reminds me of a passage in Hayek’s “The Gold Problem” (1937; published in Good Money, Part I). In this article, Hayek discusses the problems of an over- and under-supply of gold, as experienced between ~1920–1937. He spends part of the article blaming problems with the gold standard on bad monetary policy,1 and later he discusses some proposed solutions to the problem of gold, the price level, and nominal shocks. One of these is by Lionel Robbins, who proposes that government hoard (save) surplus revenue during booms, and use these savings to fund counter-cyclical policy. Hayek gives an interesting response,

There remains a third option, however, which has a good deal in its favour at least theoretically and is likely to play an important role in the future discussions of this problem. We refer to Lionel Robbins’s recent ingenious proposal that excess gold be used to finance the government’s anti-cyclical and anti-recession policy. The traditional methods for financing such measures unfortunately have the built-in tendency to counteract their effectiveness. There is nothing wrong with the basic concept to set aside reserves to be used for unemployment compensation and public works in times of recession during the preceding period of prosperity. The trouble arises when these reserves are invested in the usual fashion and then must be withdrawn from the market as needed during the slump. The forced accumulation of additional capital in boom times further stimulates the overheated investment activities, while the liquidation of this fund during the slump, when it is needed, deprives the economy of funds at the very time that they should be supporting investment activity. A rational investment policy would require such funds, if they are raised by tax revenues, to be truly hoarded, so as to be available as a ‘war chest’ for deployment in times of recession. The relative reduction of the money supply during periods of economic expansion and its relative increase when the economy is in a decline would counteract the inverse changes in the velocity of money that regularly manifest themselves in these two phases.

— p. 183.



  1. This article ties Hayek’s business cycle theory together with the theory that a mismanagement of the gold standard caused the Great Depression, by forcing countries to tighten the money supply and exacerbate (or cause) a secondary deflation.

Nominal and Real Wages During the Great Depression

Paul Krugman, following Keynes, argues that downward wage adjustments are not conducive to recovery, because real wage rates don’t tend to adjust even when nominal wages are cut. The theory is that output prices will move in line with nominal wages. To justify this theory, Keynes assumes that the most significant determinant of marginal cost are wage costs (wages multiplied by labor, or wL). I want to look into this, but I just wanted to quickly put up some evidence that contradicts the Krugman–Keynes story.

Unfortunately, the data that is most accessible to me comes from Krugman’s “favorite” book on the Great Depression, Rothbard’s America’s Great Depression. Rothbard, in turn, gets his data from Sol Shaviro, specifically from his unpublished manuscript, “Wages and Payroll During the Depression.” Here is that data,

Nominal and Real Wages 1929-33

Keynes’ theory about the relationship between real wages and nominal wage cuts assume that nominal wage cuts will have no effect on real wages. The data from December 1931 to March 1933 shows that this simply isn’t true. Nominal wage cuts correlate with falls in real wages, even if these rates of change are not symmetrical.

I know there is a more complete set of data out there (I can find real wages for the same years in Robert A. Margo’s “Employment and Unemployment in the 1930s,” but he doesn’t list nominal wages [or the wholesale price index he uses those deflate nominal wages] — and I have been having some trouble finding that data online); maybe I’m not looking hard enough.

To me, in any case, the premise that nominal wage cuts will not lead to real wage decreases, because the price of output will adjust to their new marginal costs (which must be almost entirely determined by labor costs), seems to fly against both reality and what we predict in the opposite case. If nominal wages were to rise, say because of labor union involvement, what we predict is unemployment, because the real wage will rise above the market clearing wage. The assumption is that the price level will stay the same. But, what we would predict based on Keynes’ assumption is that nominal wage increases will raise the price level, because the price of output will rise to reflect increases in marginal cost. Admittedly, this is consistent with, say, a wage-push theory of inflation, but it’s not what (I thought) most economists would expect to happen (and I thought most economists, except some of those on the margins, believe that inflation is “always and everywhere a monetary phenomenon” — an increase in M or a fall in the demand for money).

Krugman at His Strangest

This is a strange argument by Krugman.

(N.B. I agree with his main assumption — we are in a demand-side recession. I also think that economists who don’t recognize the logic of the demand-side recession should re-think their position. I’m not disputing this element of Krugman’s argument.)

It needs to be recognized that there are probably equally as many conservatives (at least, people Krugman would see as conservatives) who share the same general explanation for the Great Recession as Krugman. These are economists like Scott Sumner, Steven Horwitz, George Selgin, Larry White, Milton Friedman, et cetera. However, imagine the only blog you read is Krugman’s; you would have no reason to think that conservative monetary disequilibrium economists exist.

The impression I get as a reader of Krugman’s blog post is that if you agree with the general demand-side approach, then the rest of it naturally follows. That is, anyone who believes that, to one degree or another, unemployment benefits distort working incentives must think that the recession occurred because of half of the workforce was killed off in an alien invasion (oh wait, that would actually boost the economy — I kid, I kid).

Yet, there are many economists who both endorse the demand-side cause of the recession and the notion that unemployment benefits, on some margin, distort incentives to work. Note, these economists don’t necessarily think that all unemployment benefits should be cut, or even that unemployment benefits shouldn’t be extended during downturns. Friedrich Hayek, for example, supported redistributing income to guarantee a minimum standard of living. Hayek also believed in the possibility of demand shortages. It shouldn’t be controversial, in any case, that certain unemployment/welfare policies can create disincentives to work, to whatever extent — even if you think, considering other factors, that unemployment payments, on net, create a net benefit.

All of the following beliefs are consistent with each other. Unemployment benefits reduce incentives to work. There are some people, on the margin, who therefore choose not to look for work while receiving benefits (even if it only explains a tenth, or a twentieth, of the cyclical increase in unemployment). It follows that unemployment benefits don’t help reduce unemployment, quite the opposite actually. Yet, it is still a net benefit to redistribute income to society’s poorest, because there is a moral case in support. Cyclical increases in unemployment can be lengthy, and people shouldn’t starve to death because no one can afford their labor. (And, no, it’s not obvious that private charity, in the absence of public welfare, would be sufficient.)

Krugman argues that unemployment benefits boost demand, because they “put money in the hands of people likely to spend it.” But, this doesn’t follow from the demand-shortage theory of recessions. You need to make an additional assumption: the Keynesian multiplier.

In the context of the Keynesian multiplier, increasing unemployment benefits also increases investment, because, after all, someone needs to consume what is produced. The poor generally have a higher marginal propensity to consume than the wealthy; they save a smaller fraction of their income. The marginal propensity to consume is directly related with the marginal efficiency of capital (the return on investment), meaning lower consumption correlates with lower investment. Redistributing wealth from the rich to the poor therefore stimulates aggregate demand, by increasing both consumption and investment demand. But, investment and consumption require an intertemporal tradeoff, and therefore it doesn’t make much sense that current consumption drives current investment. Entrepreneurs borrow now to satisfy future consumption. There is no Keynesian multiplier. In fact, unemployment benefits increase consumption at the expense of investment, and can therefore reduce the stream of future consumption output.

So, there is a moral case for unemployment benefits, but there is no stimulus case for them. Of course, reasonable people can disagree on how large unemployment benefits should be, and for how long they should be paid. Similarly, it is reasonable to disagree on what minimum standard of living a minimum basic income should maintain. And, it follows that reasonable people can think that real world existing welfare benefits are higher than they ought to be. Of course, reasonable people should also corroborate their views with the empirical evidence, but the evidence in this case is open to interpretation.

There is, in fact, a better way to stimulate the economy (for efficiency concerns): monetary policy. And, this is exactly what many conservatives call for, and have been consistently calling for. Scott Sumner comes to mind as someone who has persistently and consistently advocated for changes to monetary policy, specifically NGDP targeting. Krugman, of course, agrees. He wrote a well-known paper on how monetary policy (increasing inflation expectations) can solve a demand shortage, even at the zero lower bound. But, it’s funny that he doesn’t mention this when he’s talking about conservatives who reject unemployment benefits and how unemployment benefits can stimulate the economy.

Separate quibble: recessions and the capital stock

Krugman asks, “Did termites eat part of the capital stock?” He answers in the negative. He’s wrong, and he might even be contradicting himself.

To keep it simple, look at the housing market. During the housing boom, there was quick growth in the construction industry, as new houses began to built in certain parts of the country. For whatever reason, it turned out that this investment was unsustainable, because consumers’ incomes fell (their budget constraints shifted to the left) and the demand for housing fell. The capital that was used to produce this excess supply of housing was misallocated; it was consumed at a loss. The capital stock shrinks. Potential GDP falls.

Of course, shocks to the capital stock can’t explain all of the movement in cyclical unemployment. If wages were perfectly flexible, the labor market would clear and there would no involuntary unemployment (apart from that produced by supply-side frictions). Cyclical unemployment can only be explained by non-clearing labor markets. Yes, wage floors (i.e. the minimum wage) are also responsible for non-clearing labor markets, but a large fraction of the unemployed are more productive than minimum-wage earners. These people are unemployed because of a shortage of money. Thus, monetary policy.

Does it Matter if the Government Borrows at Low Cost?

Every once in a while, someone — for example, John Aziz and Brad DeLong — in the economics blogosphere argues that low borrowing costs suggest that we should increase government spending, rather than decrease it. But, do interest rates on government bonds really signal to us the optimal quantity supplied of government investment? Not only does it not, but they could in fact distort the desirability of government spending, hiding the true costs of the public allocation of goods.

Suppose a recession is characterized by a demand shortage. This shortage is caused by an excess demand for money, and the implication is that there is a stock of idle resources, because they’re priced out of the process of exchange. There are three general solutions,

  1. Wait for the demand shortage to be corrected by a falling price level. If the demand for money increases, the value of other goods relative to money must fall, implying that their prices must fall, as well. This is what we mean by a falling price level. If the price level falls, the real value of individuals’ cash balances will increase, and the shortage of money will end;
  2. Use the central bank to increase the money supply to resolve the money shortage. Alternative monetary policies include NGDP targeting, high inflation targeting, et cetera. The idea behind monetary policy is to end the money shortage, but without directly allocating “real” resources;
  3. Fiscal policy, which is the direct allocation of resources by the public sector.

The desirability of fiscal policy increases during a demand shortage if (a) prices are sticky and (b) interest rates are approaching or at a lower bound. This lower bound — now usually considered to be zero — matters, because at the bound the demand for money becomes perfectly elastic. At the lower bound, the return to lending is too low to induce savers to lend (disregarding the costs to holding money) and they prefer to hold newly issued high-powered money. This includes banks, which prefer to hold new reserves over lending them. This is what Keynesians mean when they argue that “conventional” monetary policy loses traction at the lower zero bound and fiscal policy becomes attractive.

The case for fiscal policy is strong, even if I don’t find it persuasive, but does the public cost of borrowing have anything to do with it?

(a) Consider that public investments are not supposed to return a visible profit. In fact, we opt for public investment precisely when the typical market signals that attract private investment don’t work, because, for example, many of the benefits are not internalized by the investor. The New Keynesian case for fiscal policy is basically that there is a large macro market failure, which is that many worthwhile exchanges are not taking place because of a money shortage and the zero lower bound. Even if the rate of interest on government bonds increased during the recovery — like in Spain and Greece —, we could make a New Keynesian argument in favor of countercyclical fiscal policy. The direction of the movement in the rate of interest wouldn’t play a role in the argument, because we wouldn’t expect the government to be able to internalize all of the benefits of the investment — a cost/benefit analysis would be misleading;

(b) Think about what a falling interest rate on a mortgage backed security, not necessarily in the context of a recession, suggests about that asset. Assume that time preference is fixed. A falling interest rate suggests that the asset is becoming more liquid, which in turn implies that the asset is becoming “safer.” More people are willing to accept it at par in exchanges (which is what happened with the mortgage backed security during the housing boom, when it was used as collateral in repo markets). The rate of interest says something about the asset and about the industry behind it; namely, it suggests that the industry is profitable, because otherwise the risk of a mortgage backed security losing value would be higher. (Of course, sometimes private asset yields are misleading, but this doesn’t have to do with my argument regarding public bonds.)

When dealing with public bonds, the rate of interest is not signalling the same thing. Yes, U.S. government bonds are considered safe and they have a low corresponding rate of interest. But, this rate of interest isn’t low because government expenditure is, on average, profitable, or becoming more profitable. The cost/benefit analysis to government spending is not directly reflected in the cost of borrowing. Rather, the low cost of borrowing reflects two things. First, whether a specific investment clears the cost/benefit test has a very minor affect — at the margin — on the amount of tax revenue the government will bring in. Second, a government that controls the currency always has a last-resort when repaying debt, which means that the likelihood of default — especially on internal debt (public debt governed by U.S. law) — is very low. This latter aspect is, in a sense, a moral hazard, because it makes cost/benefit considerations irrelevant in the decision to lend.

Conceptualizing my argument through a supply and demand model might help clarify it. The supply curve represents the cost of borrowing, or the opportunity cost of lending to government. The demand curve represents the benefits received by the lenders after each marginal change in the quantity of money lent. In the private sector, “irrational exuberance” aside, there is some close relationship between this benefits schedule and the benefits schedule to the borrowers’ spending. In the public sector, however, this connection does not necessarily exist, because it’s not the profitability of the investment that informs the demand and supply curves for public debt. Thus, it’s not necessarily optimal to be at the margin where the costs of borrowing are equal to the benefits to lending.

Musing on Capital

I am writing a review of Israel Kirzner’s Competition and Entrepreneurship, to be published on this blog probably some time next week, so a short post will have to do. Besides, thinking about Kirzner’s book has me thinking about this topic.

Traditionally, the rate of interest is pivotal to the process of intertemporal coordination. When individuals save more, the rate of interest falls, individuals increase their borrowings, and demand is transferred from the late to the earlier stages (the latest state being directly prior to consumption). Perhaps the most famous critique of this view is Keynes’, who thought that the rate of interest was not a good coordinator of savings and investment. And, when this coordination process breaks down, we suffer an output gap (i.e. the business cycle). This line of reasoning has often been used to put in doubt the logic behind Austrian business cycle theory. So, the debate has hinged on the rate of interest.

I’ve long thought that the rate of interest is actually not that important. I have argued that in the boom phase it’s not a lower rate of interest that we should be looking for, but for the rate of credit expansion. I think this view is supported by Mises, in Human Action, albeit not as explicitly. More to the point, I’ve also made the argument that we should re-think the theory that it’s the rate of interest that is the key price in intertemporal coordination. It’s an argument that should be well received by Austrians. When all we need to do is solve a system of equations, the rate of interest is key to intertemporal coordination. But, this is not how coordination actually works. Kirzner introduces entrepreneurship as the main process behind coordination. The pricing process is important, in that it disciplines poor decisions and rewards the good ones, but this is an ex post role, not an ex ante one (prices have an ex ante role in optimization problems). Why haven’t we exported this concept to the theory of capital?

Of course, who am I to challenge professional Austrian economists who have thought about these issues as a career? But, consider the following passage from Hayek’s Pure Theory of Capital,

It will be evident by now that savings are actually required only at the time when, in consequence of a past diversion of input from the production of consumers’ goods to the production of capital goods, the current output of consumers’ goods is falling off.

— p. 276.

The conditions for intertemporal equilibrium are pretty tight. Savings are only actually needed when the supply of consumers’ goods begins to fall. That is, after the initial investments were made. Investment is an act of entrepreneurship, because the investor has to guess where the demand curves for different types of goods will be at the point in the future when his output will be made available, so to speak. If the rate of interest on loans is decided by the rate of savings, the interest rate will fall after the relevant investment has been made. It doesn’t make sense, then, that the rate of interest plays an ex ante role in coordination. It can only act as part of the disciplining mechanism of the pricing process.

While I haven’t read everything written on capital theory, I’ve read quite a bit, and I’ve never seen this approach adopted. Why? (Coincidentally, one book I have yet to read — and I will be doing so at some point early next year — is Kirzner’s Essays on Capital and Interest.)