Tag Archives: macro

What’s Right and Wrong With Austrian Macro?

3 Replies

“There’s something wrong with everything. In macro; not, you know, in life.” That may not be a verbatim quotation, but I remember Tyler explaining this in PhD macro I, and it has stuck with me. You don’t really understand a school of macroeconomic thought until you can dispassionately evaluate both its strengths and weaknesses. If your answer is that it has no strengths, then you don’t understand it; lots of very smart people developed the theory for a reason. If your answer is that it has no weaknesses, then you don’t understand it; lots of very smart people disbelieve the theory for a reason.

I’m writing this post on Austrian macro because the Austrian school seems to be both en vogue with and poorly understood by Tea Party types. For that matter, it is poorly understood by critics of the Tea Party. I’ll be the first to admit that I am not the most qualified person in the world to write this post. I am not an Austrian or a macroeconomist. Lots of people, including some GMU first-years who are taking the macro prelim this weekend, could do a better job than I will do. Maybe they can comment and refine the post that way. Let’s get started:

What’s right with Austrian macro?

The starting point for modern macroeconomics is what is known as dynamic stochastic general equilibrium (DSGE) models. These models vary depending on the point that the theorist is trying to make, but in the broadest class of them, there is in fact very little economics even going on. Say we start with Long and Plosser’s classic RBC model. How many goods are there in Long and Plosser? One (plus leisure). How many people are there in Long and Plosser? One! How much trade is there in Long and Plosser? Zero. Now, if what you mean by economics is intertemporal optimization in the face of random shocks, then Long and Plosser is an economic model. But as Hayek argues, “This, however, is emphatically not the economic problem which society faces.” DSGE models are poorly suited to evaluating changes in an economy with arbitrarily diverse agents with arbitrarily diverse preferences and arbitrarily diverse information sets. This critique of DSGE-style macro is part of the core of Austrian theory.

Furthermore, in the Austrian view, capital is heterogeneous and multi-specific. If you invest in a pet store, and then decide you want to convert it to a massage parlor, that is costly and time-consuming. This opens the door to malinvestment. In many RBC and New Keynesian models, capital is homogeneous, meaning that it is costless to switch from one investment into another. Kydland and Prescott explicitly assume time-to-build, but this is not the only friction in real-world investment.

According to the Austrians, production functions for the multi-specific capital are discovered over time. In virtually all RBC and New Keynesian models, production functions, the way of transforming the single type of capital into the good or (rarely) goods are specified in advance and do not change. Austrians emphasize competition as a discovery procedure. Entrepreneurs are constantly trying to find new ways to turn existing capital stocks into goods that consumers may want. This discovery procedure is obviously sensitive to policy shocks.

What is wrong with Austrian macro?

The biggest problem with the Austrian school is a legacy of the fact that much (not all!) of the theory was developed before the rational expectations revolution. Even if you think rational expectations is bogus, the fact is that many Austrian models do not explicitly state what is driving expectations. If the monetary authority inflates, everyone is tricked. This is clearly problematic. A better Austrian theory in my opinion would evolve along the lines of the Lucas islands model. I have written before that I think our current situation is one of severe model uncertainty. Some people think money is tight and others think money is loose. If that is the case, then even if people have, say, Bayesian expectations, many of them will be tricked, resulting in monetary distortions.

The other major weakness in Austrian business cycle theory is that it focuses way too much on one particular distortion, monetary policy. Now, the Austrians have an answer to this; they argue that money is one half of every single transaction in the economy, so if money is distorted, then that is a big problem. I don’t think this is true. First, monetary distortions will cause primarily intertemporal distortions. This may be problematic, but as I wrote above, one of the biggest strengths of the Austrian model is that it takes seriously the heterogeneity of goods, capital, and preferences. Focusing primarily on intertemporal investment gives away that huge gain. Austrians should be more open to examining other distortions, such as the subsidization of fixed-value financial claims (FDIC insurance, favoring debt versus equity) and industrial policy. I think Arnold Kling is onto something with his emphasis on Patterns of Sustainable Specialization and Trade.

What is misunderstood about Austrian macro?

The Austrian mantle has been claimed by the Tea Party, but very few Tea Partiers are familiar with modern Austrian scholarship. Michelle Bachmann famously takes Mises with her to the beach, but there is a great deal of Austrian theory that is post-Mises. In particular, most of the modern Austrians I have talked to are not goldbugs. They understand that the most important characteristic of money is not its store-of-value property. In general they favor rules versus discretion in monetary policy, but those rules are ones that non-Austrians can easily get behind. For instance, George Selgin writes, “Scott Sumner’s general views on macroeconomics are so much in harmony with my own that, in commenting on the present essay, I’m hard pressed to steer clear of the Scylla of fulsomeness without being drawn into a Charybdis of pettifoggery.” Furthermore, to the extent that Austrians like gold as currency, they like it because they believe it would “win” in a free-market competition against fiat currency, not because gold is special per se. A simple test would be to get rid of capital gains taxes on gold and other assets and see what wins.

Modern Austrians view policies like NGDP targeting as coming straight out of Hayek, who wrote about the importance of preventing a “secondary deflation.” Consequently, the mainstream accusation that Austrians favor no policy in the face of a financial crisis is misguided. The correct policy, according to many Austrians, is to adopt the most non-distortionary monetary policy there is, which is to keep nominal spending at the expected level. Letting spending collapse is itself a distortionary policy.

The Tea Party is populist, but it seems to be populist for the sake of populism. Austrian theory, on the other hand, is anti-elitist because it believes that neither elites nor anyone else can successfully “manage” the economy. There is consequently a certain populist interpretation of Austrianism; but the theory is not so much about giving the masses what they want as about letting a decentralized process take place. This is the main reason I am skeptical about the political adoption of Austrianism. It is being used as a rhetorical tool in a cultural dispute, not as a way of understanding the nature of the economic problems we face.

Why RBC is Awesome

Leave a reply

Paul Krugman disses RBC theory and those who study it as unscientific. I’m not an RBC theorist, but I’ll stick up for freshwater macro. Here are some reasons why RBC theory deserves more respect than Krugman gives it.

1. Suppose monetary policy is conducted so that all nominal shocks are perfectly offset. Zero percent of net shocks, shocks adjusted for changes in the quantity of money, are nominal; 100% of net shocks are real. Therefore as monetary policy improves it is the policy conclusions of the RBC literature, not the New Keynesian literature, that are relevant.

2. Empirically, I agree with Krugman and others contra pure RBC theory that money appears to be non-neutral. But why is money non-neutral? Saltwater macro offers answers to this question that are frankly absurd. Menu cost arguments rely on a fallacy of composition. Even if all businesses face costs of changing prices, unless they synchronize their price changes, there is no reason to believe that the price level as a whole is sticky; see Caplin and Spulber. If the nominal price level is not sticky, there is no reason to believe that real stickiness—efficiency wages and so on—can be the source of monetary non-neutrality. Freshwater macro has advanced the much more plausible idea that money is non-neutral because of legal restrictions on financial intermediation. Under laissez-faire, money would be neutral, and RBC would be correct.

3. Critics of RBC argue that it is hard to find shocks to real factors such as technology, particularly negative shocks, to account for recessions. However, they overlook shocks to credit, which is—wait for it—a real not a nominal factor.

4. Critics of RBC ridicule the theory by saying that according to freshwater economists, the Great Depression was the Great Vacation. But their own theory is no less ridiculous. According to saltwater economists, unemployment of over 20 percent persisted for almost a decade because people were too stubborn to accept wage cuts. Sorry, not plausible.

5. RBC makes strong assumptions, but it should be appreciated for what it is, which is an attempt to do macro as pure economics without post hoc additions to make the theory fit with measurement (which is, after all, imperfect). In contrast, other approaches exhibit a tendency toward, to coin a phrase, Macro of the Gaps. The models don’t always make this clear, but a lot of the features that do the work in saltwater theory are residuals. I am not aware, for instance, of a serious effort to give a theoretical account of the value of the fiscal multiplier. Instead, the fiscal multiplier is whatever regressions say it is. I am not against empirical work, but how convenient that regressions set the multiplier to whatever level is necessary to make the theory work. In fairness, some RBC papers empirically calibrate models, but there is more honesty about the fact that they are calibrating, not independently estimating results.

6. As Tyler used to say in Macro I, 98 percent or more of business cycles in human history were indisputably real business cycles. There was no central bank and no fiscal authority in caveman days. “My theory explains 98 percent of business cycles” seems like pretty good justification to me.

As I said before, I’m not an RBC theorist, but not everything has to be about petty tribalism. RBC is both scientific and worth studying.

What Monetary Stimulus?

10 Replies

I was chatting with a friend about the state of the economy, and I brought up how the monetary expansion that began in 2008 wasn’t actually stimulative because the Fed started paying interest on excess reserves. My friend expressed a little skepticism that excess reserves mirrored the monetary expansion one-for-one. But behold:

No, it’s not literally one-for-one, but it’s close. Non-circulating money gives me the vague impression of a tree falling in the woods with no one around to hear it.

How Much Macro Can You Explain in One Graph?

6 Replies

Macro is hard; here is an attempt to make it easier. I will try to explain as much of macroeconomics as possible with a single graph. Dynamic AS/AD is pretty good, but I think we can do some interesting things by turning to portfolio theory.

Macro is and always has been about investment. Keynes noted that investment is much more volatile than consumption, which is why his consumption function has a positive intercept and a slope of less than 1. So to explain macroeconomic fluctuations, we have to explain investment.

The graph that I’ve chosen to explain most of macro is a standard one from finance. The straight line (called the capital allocation line) shows investment options. Investors can select low-risk, low-return projects or high-risk, high-return projects. When capital markets are efficient, the line is going to be perfectly straight, but that’s not super-important under ordinary conditions. The indifference curve represents the preferences of an investor. When investors are highly risk-averse, their indifference curves are going to be relatively steep. When they are more risk-neutral, their indifference curves will be flatter. The tangency represents the portfolio selection of the investor.

We have done enough work now to shed light on the concept of animal spirits. What happens when an exogenous shock makes investors more risk-averse? Their indifference curves get steeper and they become less willing to bear risk in their portfolios. As a result a number of investment projects with high returns do not get funded, and total returns and therefore total output in the economy go down. Of course, it need not be an exogenous shock; regime uncertainty will also make investors more risk-averse, decrease total risk-taking, and decrease expected output.

That’s it for the indifference curve; let’s now turn to the capital allocation line. The slope of the line measures the incentive to bear a marginal unit of risk. If there is not enough risk-taking in the economy, the government can try to stimulate risk-taking by increasing the slope of the capital allocation line. It can do this in two ways, either by lowering the return to low-risk projects or by raising the return to high-risk projects.

Lowering the return to low-risk projects is usually called monetary stimulus. One very low risk project is to hold cash or Treasuries. Monetary stimulus lowers the return to holding cash by creating inflation. Since the real return to cash is lower, people substitute Treasuries for cash, lowering the real rate of return on Treasuries. Other low-risk investment projects, which are close substitutes for cash and Treasuries, will also experience decreases in real returns in proportion to how close of a substitute they are—that is, how riskless they are. As a result, the slope of the capital allocation line increases, and investors take on more risk than they would without the stimulus.

Fiscal stimulus attempts to raise the return to high-risk projects. The government subsidizes projects that are not close substitutes for holding cash and Treasuries. Like monetary stimulus, it makes the capital allocation line steeper, though it works from the other end of the line.

We have good reason to think that monetary stimulus is superior to fiscal stimulus. Under monetary stimulus, the projects that bear the additional risk are selected by investors. Under fiscal stimulus, the government selects the additional risk-bearing projects. If you believe, as I do, that investors are better at selecting the investment projects to bear the marginal risk than the government, you should prefer monetary to fiscal stimulus.

What happens when the government, through monetary or fiscal stimulus, makes the capital allocation line steeper than it ought to be? Investors take on lots of risk. This can increase total returns and total output, but if the risks that investors are taking are correlated, then you can have a quasi-Austrian boom/bust cycle. Investors, though individually rational, collectively become irrationally exuberant because the price of risk is wrong. On expectation, returns are higher because of the additional risk. But every so often the risk goes badly and output falls. Any continuous policy of monetary or fiscal stimulus is going to increase the volatility of the economy.

Long-term economic growth affects the level of the capital allocation line. A fast-growing economy is going to have higher returns for both low- and high-risk projects than a stagnant economy. When an economy grows slowly enough or shrinks, the capital allocation line may go into negative real return space. When the line needs to go into negative real return space but it cannot due to, say, deflation, the marginal return to risk-bearing on low-risk projects is going to be zero. This is a liquidity trap.

I think we’ve gotten pretty far with just the one graph. Feel free to extend the analysis in the comments. I will note that nothing in this post relies on price stickiness or monetary misperceptions. It’s a real model, though of course it has a role for monetary policy in affecting the real return to holding cash. Nevertheless we are able derive some Keynesian, Monetarist, and Austrianish ideas. Many, though not all, of the ideas in this post were liberally borrowed from Tyler’s Risk and Business Cycles, which is highly recommended, though I have not read it in a while, so any errors in this post are probably mine and not Tyler’s.

Risk, Fat Tails, and Business Cycles

6 Replies

Tyler Cowen offers his non-Keynesian take on the recession, applying the theory he lays out in Risk and Business Cycles (recommended for all economics graduate students, but master Snowdon and Vane first). I agree with his arguments, but I want to add what I think is a missing ingredient in his theory: fat tails.

David Levy is the source of my appreciation for fat tails. His graduate Econometrics 1 class is basically a class on robust estimation. He makes his students do loads of exploratory data analysis—we generated data with different error distributions and then evaluated how various techniques performed at estimating the true parameters, which we knew because we generated the data. My favorite distribution is the Cauchy distribution. It is fascinating because to the naked eye it looks a lot like the normal distribution, but the techniques that work for normally distributed errors are very inefficient for Cauchy-distributed errors.

I think that real-world macroeconomic errors are distributed more like a Cauchy distribution than like a normal distribution. They have fat tails. But humans do not find Cauchy-distributed errors intuitive. Most of the errors we deal with in our ordinary lives are distributed normally. We make a cognitive error when we confront fat tails.

The implication for Cowen’s theory is this. People invest and consume thinking the world is less volatile than it is. There is a series of years in which the errors are in the main part of the distribution. People infer that the world is pretty stable. Then the eight-sigma event happens. People realize that the world is volatile and they have inadvertently been taking more risk than they intended to take; they exit risky investments and move to safer ones. Cowen’s theory takes over.

Risk and Business Cycles explicitly adopts a rational expectations assumption, and I am positing a systematic error. I think that my position is defensible. I think RE can be a useful methodological tool, but one has to recognize its limitations. RE is likely to closely describe reality when feedback is reliable. I have no problem with adopting mean-zero first-order errors in macroeconomics, but I think there is good reason to believe that people’s expectations about the shape of the error distribution is not subject to good feedback. It is entirely possible that 19 out of 20 years, the people who assume a normal distribution will outperform those who assume a fat-tailed distribution. The other years, government intervention may insulate those who feel the pain of their big mistakes.

The past few decades are sometimes called The Great Moderation. My hypothesis supports skepticism that anything significant changed during this period. The world has always been volatile; we just didn’t realize it.

Monetary Confusion

3 Replies

Scott Sumner has drawn a lot of attention for his peculiar view that monetary policy was tight in 2008, and that this tightness was the cause of the current recession and the recent financial crisis. I think his argument is at least somewhat persuasive; I don’t know if (or think that) it was the whole problem, but it does seem to me that his monetary proposal would have at least decreased the severity of the recession.

What intrigues me, however, are the implications Sumner’s view has for macro and monetary theory. Sumner is arguing that monetary policy was tight in 2008, while most other economists would argue that it was loose. In a recent post, Sumner suggests that there is no persusasive metric of monetary stance. Nominal interest rates can signal monetary stance or inflation expectations. Real interest rates were high in late 2008, but most economists think monetary policy was loose. Other indicators have their own problems.

If even economists disagree about whether money is tight or loose, what hope do ordinary producers and consumers have of preemptively adjusting to monetary policy? If people are genuinely confused about the stance of the monetary authority, this breathes new life into a class of economic models that the profession has discarded and ignored in recent decades, epitomized by the Lucas-Islands model. In the Lucas story, as I remember it, producers observe changes in demand for their products and change their production, factoring in all expected changes in inflation. If inflation is expected, then producers who observe increased demand for their products won’t be tricked into increasing production—they will realize that only nominal, and not real, demand has increased. If inflation is unexpected, producers will think the observed increase in demand is real, not just nominal, and increase production accordingly. The model therefore establishes a link between unanticipated inflation and the real economy.

The Lucas-Islands model has fallen out of favor because of the simple criticism that it is trivially possible to observe the open-market operations of central banks. On any given day, it is easy to determine, say, the monetary base, or whether the Fed is buying or selling treasuries. Therefore, the story goes, no inflation should be unanticipated. Yet we observe in reality that monetary policy seems to have an effect on the real economy; therefore the Lucas model cannot be correct.

If, however, the actions of the central bank are not sufficient to determine the monetary stance, the Lucas model (or other monetary confusion models) could still be accurate. One would have to assume that errors are clustered, but that is compatible with rational expectations. Throw in heterogeneous K and you are most of the way to Austrian business cycle theory. The logical conclusion is that economists ought to reexamine these macro models or incorporate monetary confusion into newer DSGE models (although they might then have to be renamed DSGD models). Unless someone is able to convince Sumner that monetary policy was loose in 2008 after all, it seems that taking monetary confusion more seriously is a promising way forward for macro.