The Leamer Indicators
By Arnold Kling
Now, I am getting to the part of Macroeconomic Patterns and Stories where Leamer gets very conventional. Economic activity equals spending, monetary policy equals the Fed Funds rate, and all the fiscal and monetary dials work as they do in textbooks.
Writing in 2007 and very early in 2008, he sees the expansionary monetary policy of 2001-2004 as a huge mistake, because housing was already strong to begin with. He was not expecting the severity of the 2008-2009 recession, but my guess is that after the fact he could rationalize that in addition to too many houses started during the boom, we had too much house price inflation, leading to too much paper wealth creation, leading to too much consumer spending that had to be cut back when prices crashed.
I’ll put the rest of this post below the fold.In chapter 12, Leamer says that of the ten postwar recessions, eight of them began with drops in big-ticket consumer purchases, usually houses, but often including cars and other consumer durables. Of the other two recessions, one is the Dotcom recession. The other recession he attributes to the decline in defense spending at the end of the Korean War.
The 1953 DOD downturn was a great reverse test of “Keynesian” reliance on government spending
He does not discuss the rapid decline in defense spending after the second World War, but a surge in consumer spending at that time might be explained as coming from a reduction in the extraordinarily high saving rate that prevailed during the war. On a further Keynesian note, he argues that the defense buildups for both the Korean War and the Vietnam War served to offset declines in housing starts that otherwise might have produced recessions.
On the housing cycle, Leamer writes,
Both in manias and the depression, demand is upward sloping not downward sloping. When the price is rising in a mania, a higher price suggests even higher prices later on, and buyers rush in before it is too late…When the price is falling during the depression, a price cut suggests more cuts are on the way, and buyers decide to wait
This is a story of momentum on the demand side. It would be interesting to see how much of the momentum in employment is accounted for by the construction industry.
In chapter 15, Leamer attributes to the Fed the power to make or break housing booms. On the overall business cycle, he writes
look at the behavior of interest rates preceding the cycle peak. In every case, interest rates were rising in the two-year period before the recessions began. That makes me imagine that those rising rates choked off growth and tipped the economy into recession.
How does the Fed do it, when it affects short-term rates and not long-term rates? Leamer writes,
if the yield curve flattens…banks do not make intermediation profits and they must…carefully identify borrowers with low default risk…loan approval can get much more difficult…it is called a “credit crunch,” which can put a big crimp in housing sales.
Actually, until the 1980’s, you do not need such a complicated story. There were ceilings on the interest rates that savings and loans could pay on deposits. When short-term interest rates rose, deposit inflows stopped or even went negative, and so the thrifts had to curtail mortgage lending.
That was when I was an economist at the Fed, from 1980 through 1986. There was a lot discussion about the possibility that lifting the ceilings on deposit interest rates would fundamentally change the way that monetary policy affected the economy. The thinking was that we were in the process of changing from credit rationing by availability (because deposit interest ceilings would take funds away from savings and loans) to credit rationing by price (because now savings and loans would have funds, but at higher interest rates). My point is that we saw it as a significant structural shift at the time, and yet Leamer can tell a story from 1948 at least through the 1990’s without noticing any such structural shift.
In any case, Leamer see housing as the main factor in economic fluctuations, and he sees the Fed able to influence housing. This leads him to advocate an approach to monetary policy.
The Leamer Rule has the Fed Funds rate depending on deviations of a lont moving average of inflation from the target inflation rate (2%), deviations of housing starts from normal (about 1.5 million per year), and the direction that housing starts are moving over the last year. Inflation determines the long-run movement in rates and housing the short-run movement.
Standard approaches to monetary policy, such as the Taylor Rule, use inflation and unemployment are the two main indicators for when to change the Federal Funds rate. What Leamer is suggesting is replacing the unemployment rate with housing starts. If nothing else, the benefit of this is that housing starts are a leading indicator and the unemployment rate is a lagging indicator.
I do not believe that Leamer actually writes down a quantitative rule in his book (maybe he has one elsewhere). So instead I call housing starts and inflation the Leamer indicators.
Leamer has a chart pointing to where his indicators would have suggested tightening monetary policy–he uses the expression “apply the breaks” (sic). Historically, using housing starts would have made the biggest difference in 1971, when housing starts were approaching 2 million and on their way to record levels. In restrospect, it would have been a really good idea to tighten monetary policy at that point. Instead, President Nixon and Fed Chairman Arthur Burns tried the great Keynesian experiment of using wage and price controls to suppress inflation while running a loose monetary policy. The result was a boom in 1972 followed by a horrible recession in 1973-75.
Because housing starts rebounded and were about 1.8 million by 1977, Leamer thinks the Fed should have tightened then. Instead, it was not until 1980 and the appointment of Paul Volcker that the tightening occurred, by which time inflation had risen to double-digit levels and housing starts were already starting to decline.
The next point at which housing starts called for tightening according to Leamer was the onset of the Reagan recovery in 1984. However, using other indicators, tight money at that point probably would have been a mistake. Inflation was trending down and unemployment was high. In hindsight, it is hard for me to see any harm caused by the failure to raise interest rates in 1984.
About recent policy, Leamer writes,
the Fed can affect the timing of the building of new homes, but cannot much affect the total. Thus, after a normal housing-led recession, interest rate cuts are especially effective because building can be moved forward in time–capturing the sales not made during the recession. But…in 2001 in which housing plowed ahead with little notice of the softness in the economy, subsequently low levels of interest rates inevitably transfer sales backward in time, encouraging sales that might otherwise have been made years later.
Thus, Leamer’s housing start indicator would have suggested tightening monetary policy as early as 2002. Given what eventually transpired, it is hard to argue with the notion that it would have been better to choke off the housing boom back then. But for the economy as whole, that would have amounted to taking away the punch bowl before there was even a party. (That would also have been the case in 1984).
Again, Leamer is telling a fairly conventional macro story. Economic fluctuations consist of changes in spending, typically related to housing (I say “related to” to include the wealth effect of the housing bubble). The Fed has a profound effect on housing, even though it only controls the short-term interest rate. Fiscal stimulus works, as shown by the Korean War buildup and slowdown as well as the Vietnam buildup.
I could say that Leamer found a spending story because that is what he was looking at. He looks at the components of spending and asks how they behave during recessions. That leads him to housing and consumer durables as major culprits. Where does that leave the Recalculation Story, in which a recession occurs when unsustainable patterns of specialization and trade break down, and the economy has to find new, sustainable patterns? Some possibilities:
1. The Recalculation Story does not apply at all. You mostly have housing cycles, and what we just had was the mother of all housing cycles.
2. The Recalculation Story does not apply to postwar twentieth century cycles, but it applies to the Dotcom recession and the current slump. It also applies to the Great Depression.
3. The Recalculation Story can be applied to all recessions. It just turns out that in the 20th century postwar recessions, unsustainable patterns of specialization and trade were short-term phenomena related to housing and consumer durables.
I want to lean toward (2), but I would like to see more analysis of employment behavior. I think that the more that a recovery consists of workers returning to the same sorts of jobs that they held before the recession, the more I like the spending story and the less I like the Recalculation Story. I see the Great Depression as a Recalculation Story because of the huge shifts that took place between 1930 and 1950. Agricultural employment falls off, and white collar employment picks up, as the economy adapts to motorized transportation. I see the 21st-century economy as a Recalculation Story. Again, blue-collar work is on a downward trend, interrupted somewhat by the housing bubble. Meanwhile, the pattern of white-collar employment needs to be reconfigured, as the economy adapts to the Internet.