American Macroeconomics, 1960-2012

1. The Major Viewpoints (see my previous post)
2. Targets, Instruments, Indicators, Patterns, and Stories (introduces key macroeconomic data and standard interpretations for 1960-2007; Benjamin Friedman meets Edward Leamer)
3. The Hicks-Friedman 70’s textbook synthesis. IS-LM and AS-AD in all their glory (but I hope to do this without diagrams).
4. Why macroeconomists would be better off as historians than as physicists. Causal density makes it unwise to try to draw firm conclusions. At a theoretical level, The textbook diagrams make it appear as if macroeconomic variables are stable and path-independent. Yet the Great Stagflation and the Financial Crisis Aftermath seem to represent “phase changes” or “regime shifts,” at least compared to prior experience and expectations, and macroeconomic variables exhibit obvious path-dependency. Turning from theory to empirics, we do not have controlled experiments available to us. Statistically, we cannot view macroeconomic data as consisting of repetitive trials coming from a common environment.
5. How I experienced the 1970s convulsion. Follow me as I enter Swarthmore college in 1971, learn 1960s Keynesian macro, and proceed to see what I have learned fall apart in just a few years.
6. The Many Descendants of Robert E. Lucas, Jr. and Stanley Fischer. How the macroeconomics profession became narrow, inbred, and retarded.
7. The Dirty Laundry Hidden in the 1970s Textbook Model. Problems include: the unspecified connection between short-term nominal interest rates and long-term real rates; the hypothesis of sticky nominal wages is asked to carry a very heavy load; money has close substitutes both as an asset and as a transaction medium; the paradoxical relationship between money demand, the price level, and inflation
8. The Financial Crisis and its Aftermath. Are any of the pre-2000 viewpoints adequate? What about the post-2000 viewpoints?

Here is a table to organize the various viewpoints (K means Keynesian, C means Classical, S means a synthesis):

Notes: I date classical to the publication of The General Theory, because Keynes defined classical as a contrast to his views. I date 1960’s Keynesian to the Samuelson-Solow paper on the Phillips Curve. I date the 1970s textbook synthesis, which might be called the Friedman-Hicks synthesis, to the publication of Milton Friedman’s address to the American Economic Association (delivered in 1967, published in 1968). I date the general disequilibrium model to the publication of the Barro-Grossman book. I date the New Classical to Lucas’ JET paper. I date New Keynesian to the John Taylor and Stan Fischer papers. I date populist supply-side to the 1980 election campaign. I date academic supply-side (RBC) to the Kydland and Prescott 1982 paper. I date the liquidity trap to Krugman’s 1998 Brookings paper (yes, I know that Keynes was there first). I date the Minsky Moment to the 2011 edition of Kindleberger’s book (that’s also open to challenge). I date PSST to my paper in Capitalism and Society. I date Market Monetarism to Marcus Nunes’ book.

Note that the general disequilibrium, populist supply-side, and academic supply-side never really had much effect on the consensus.

One way to think of the history is that classical economics could not account for the Great Depression, and so we ended up with 1960s Keynesian as the consensus. But 1960s Keynesian could not account for the 1970s Great Stagflation, and so we ended up with the textbook synthesis as the consensus at the undergraduate level and the New Keynesian as the consensus at the graduate level. Policy thinking in Washington remained somewhere in between 1960s Keynesian and textbook synthesis, with some populist supply-side thrown in.

None of the macroeconomic viewpoints of the 1970s or 1980s could account for the aftermath of the Financial Crisis of 2008, so we had the last four emerge as contenders (I know it’s brave to put PSST in there, but it is going to be my book, after all).

I got 20,000 words written, but now I want to start over. Current thoughts:

1. A concise, selective history of macroeconomic thought from 1960-2012. Not a survey or scholarly reference work. By history of thought, I mean to capture the path-dependence of the ideas. Starting in 1960 lets me skip over a whole lot of the discussion that preceded 1960. “What did Keynes mean?” is a well-squeezed orange that I do not want to touch.

2. Organized in two dimensions (three if you count time). One dimension is by school of thought, covering: classical, 60’s Keynesian, Friedmanite, 70’s textbook, general disequilibrium (Clower, Barro-Grossman) New Classical (rational expectations, real business cycle), New Keynesian (rational expectations with sticky wages/prices), Minsky, PSST.

3. The other dimension is by major issue.

Unused Resources–are they a temporary aberration that markets will quickly eliminate? a manifestation of effective demand failure? other?

Wage stickiness–are wages sticky in real terms? in nominal terms? How central is this to explaining macroeconomic phenomena?

Macroeconomic data–is it well behaved? If not, what are the important issues: path dependence? phase changes/regime shifts? structural change? singular events (policy shocks, exogenous shocks)? Can we cut through the causal density in order to test theories?

Fiscal policy–is it an important tool for macroeconomic stabilization?

interest rates–one interest rate or multiple interest rate?, real vs. nominal, ex ante vs. ex post

Money–how unique is it? how do substitution possibilities in terms of assets and transaction media affect the link between central bank actions and the economy?

Credit–do we need to understand credit markets, including credit rationing and changes in risk tolerance, in order to understand macroeconomic behavior?

Mark Perry reports,

Employment in the durable goods sector for motor vehicles and parts increased in July to 818,000, the highest employment level in that sector since October 2008. Over the last quarter, employment in the US auto industry has increased by 10.8% at an annual rate.

Jeffrey Sparshott of the WSJ blog reports,

Residential and specialty trade contractors — home builders — added 6,300 jobs in July.

Autos and new home construction used to be the big cyclical sectors of the economy. If I take 10 percent of 818,000, I get 82,000. If I multiply 6300 by 12, I get 75,000. So at an annual rate these erstwhile behemoths are adding 157,000 jobs, which amounts to rounding error in a work force of 140 million.

One of the ways in which my macro view differs most from that of the textbooks is that I believe that it is very hard to identify repeatable patterns in an economy that undergoes such important evolution. Developments that probably change the way the economy reacts to fiscal and monetary policy include:

–deregulation of deposit interest ceilings in the 1980s , which reduced the credit-rationing impact of higher interest rates

–change in the labor force participation behavior of men and women

–big drop in the proportion of work force employed as manufacturing production workers

–big increase in the use of credit cards, Internet banking and ATM machines; we can go months without using our checkbook).

–big shift toward education and health care, or what Nick and I called The New Commanding Heights

In this section, I want to introduce four possibilities for the effectiveness of policy instruments in helping to reach a target for the unemployment rate. In the classical model, neither instrument is effective. In the crude Keynesian model, deficit spending is effective, but changes in the Fed Funds rate are not. In the textbook AS-AD model, both instruments are effective. In the crude monetarist model, changes in the Fed Funds rate are effective, but changes deficit spending is not. We can summarize this in a table.

…First, a purely classical theory would say that employment is determined by productivity and workers’ preferences for labor and leisure… There is nothing left for macroeconomic policy to do, other than determine the price level…

In the crude Keynesian model, the budget deficit affects real output, because a larger deficit raises the interest rate, which increases the velocity of money…However, the Fed Funds rate has no effect on output. A decline in the Fed Funds rate raises the money supply, but this leads to an offsetting decline in the velocity of money…

In the textbook model, both instruments affect real GDP. As in the crude Keynesian model, a larger budget deficit raises the interest rate and increases the velocity of money. However, unlike in the crude Keynesian model, in the textbook AS-AD model when the Fed Funds rate goes down and consequently the money supply increases, there is not an offsetting decline in the velocity of money…

Finally, we have a variation on textbook AS-AD called the crude monetarist model…Monetary policy is powerful, because with velocity fixed, aggregate demand depends entirely on the money supply. Since the Fed can raise the money supply by lowering the Fed funds rate, and conversely, the Fed controls aggregate demand. However, with velocity fixed, there is nothing that fiscal policy can affect. An increase in interest rates does not increase velocity. Instead, it leads to reduced investment spending by businesses, offsetting the increased spending by consumers and government. Such an offset is called “crowding out.”

As I work on my book, I am trying to come up with a way to explain the classical economic outlook. Here is one approach. Comments welcome.

Here is an imaginary dialog between a classical economist (C) and Keynesian economist (K).

K: See that man, Uri, sitting on the bench over there? He is involuntarily unemployed.

C: How do you know that? Do you know his reservation wage? That is, do you know the lowest wage that he would accept to go to work? Do you know what his best offer has been?

K: Yes. He won’t work for less than $12 and hour, and his best offer has been $11.50

C: So he is not really unemployed. He has withdrawn from the labor force, because he can’t find a job that will pay him what he wants.

K: No, according to the Department of Labor, as long as he is looking for work, he is unemployed. Besides, in his last job, he earned $14 an hour and what he produced was worth $15 an hour. But when the economy went into a slump, the demand at his firm fell, and he was laid off. His problem is that there is a lack of effective demand.

C: I’m not sure what ‘effective demand’ means, but ok. What should Uri be doing instead of sitting on the bench?

K: He could be digging a ditch for the government.

C: But he’d rather be sitting on the bench. Why should he dig the ditch?

K: The government can pay him to dig the ditch. They can pay him $12 an hour.

C: If his ditch-digging is worth $12 an hour, that’s fine. The taxpayers should be happy to pay Uri to dig a ditch if it’s a worthwhile use of his time.

K: Actually, the ditch is not worth so much. Let’s say his ditch-digging is worth only $5 an hour. But this way, he’s working instead of sitting on a bench, and as taxpayers we benefit from the ditch.

C: No! As taxpayers, we pay $12 and hour for ditch-digging that is worth only $5 to us. That makes us worse off.

K: Would you rather pay unemployment benefits of $8 an hour and get nothing?

C: No….But if we are going to redistribute income to Uri, why not encourage him to take the offer for $11.50 and pay him just $.50 an hour as a subsidy to do that?

K: Hmmm. Not such a bad idea. But the ditch-digging puts more spending into the economy.

C: No it doesn’t. You give $12 to Uri to spend, but that $12 comes from those of us who pay taxes, and now we have $12 less to spend. It’s just a transfer.

K: But we’re not going to raise taxes. We are going to borrow the money to pay Uri to dig the ditch.

C: In that case, the borrowing is going to use up saving that otherwise would have been used to build homes or expand businesses.

K: No. Households and businesses do not want to spend any more. The savings would have sat idle. We need the government to spend those savings, because no one else will.

C: Really? You seem to think that we can have an excess of savings without driving down interest rates. I don’t see how that can happen.

K: That’s a discussion for another day.

As I attempt to write my macro book, I keep Edward Leamer’s Macroeconomic Patterns and Stories close by. It really is one of those books that I have to read many times in order to absorb its insights.

I’m looking at chapter 15, “In Search of Recession Causes.” He writes,

give the banks a steep yield curve, falling long-term rates of interest, rising incomes of loan applicants and housing appreciation that makes loans self-collateralizing, and the banks will be very happy, indeed, and will compete intensely to find someone who wants a mortgage loan…

But if the yield curve flattens, or if overall income growth slows down or if home price appreciation stops, banks do not make intermediation profits and they must perform a different function–they must carefully identify borrowers with low default risk…it is called a “credit crunch,” which can put a big crimp in housing sales.

Be sure you understand what a credit crunch is. In a normal market, even very risky borrowers…have access to credit, if they are prepared to pay the interest rate premium…In an exuberant market, the lending standards can get very relaxed. But during a credit crunch, many borrowers simply are shut out of the market and denied credit.

In a prior post, I mentioned his credit-crunch model. As an expansion matures, long-term interest rates are going up, because of inflation fears. But the Fed is particularly worried, and it raises the Fed Funds rate relative to the long-term rate. This raises banks’ cost of funds. That, according to Leamer, reduces banks’ willingness to supply loans. But what he does not explain, it seems to me, is why the reduced willingness to supply loans takes the form of credit rationing rather than just higher interest rates on bank loans.

Here are my comments on the non-price rationing story.

1. I find it plausible that non-price rationing would cause more economic disruption than price adjustment. I believe we learned that from our experience with oil price controls and gas rationing in the 1970s. Gas lines are much worse than higher gas prices.

2. With bank lending, rationing by price might result in non-price rationing. That is, as you raise the interest rate, you find fewer borrowers who are likely to be able to make the payments on a mortgage. So perhaps the distinction between rationing by price and non-price rationing is less applicable to something like mortgage lending.

3. Prior to 1980, we had interest-rate ceiling on deposits at banks and thrifts. What this meant was that when interest rates rose, non-price rationing took place, as banks were unable to raise rates, so they could not keep depositors from fleeing. This in turn meant that mortgage credit was curtailed. So I understand how the Leamer applied back then.

4. But after 1980, we had the “atmosphere of deregulation,” which led to the banks and thrifts being able to pay a market rate to depositors. Should we have had the same type of credit crunches? I think not. And perhaps we did not. One can argue that after 1980, the relationship between the Fed Funds rate and the unemployment rate became a little more loosey-goosey. You still see the Fed able to raise the unemployment rate by raising the Fed Funds rate, but now it’s taking years rather than months for higher rates to stop the downtrend in unemployment, and there are a couple of periods (1984-ish; 1995-ish) in which tightening does not raise unemployment at all.

5. We could describe 2003-2007 as a credit anti-crunch (loosening mortgage standards) followed by a credit crunch (tightening mortgage standards by a lot). In both the anti-crunch and the crunch, government pressure played an exacerbating role, to say the least.

6. According to the Leamer model, quantitative easing should be contractionary. That is, if you think that banks ration credit when long rates are low relative to short rates, then what you want to do is let long rates rise, so that banks will lend more. I doubt that this is the right way to think about quantitative easing. Unless you have a story like (3) above, I don’t think you can say that an inverted yield curve will cause banks to ration credit.

To make a long story short, I believe in credit crunches. Prior to 1980, they were caused by the interaction of regulations with Fed tightening. And the financial crisis looks like an anti-crunch followed by a crunch. All of these crunches affected housing and consumer durables, and these are important sectors in the business cycle. However, I do not believe that, in the absence of regulatory distortions, an inverted yield curve causes a credit crunch.

Finally, I do not think that the whipsaw in the demand for housing and consumer durables is the big story of this decade. I think that the big story is structural change.