Tuesday, December 28, 2010

Money Expenditures Targeting and "Leaning Against the Wind."

Luigi Zingales has an interesting paper about the Efficient Markets Hypothesis. I found it all very sensible, until his argument regarding central banking. He argued that central bankers should "lean against the wind," when they observe a speculative bubble.

While a price earnings ratio of twenty-five might not be irrationally high, it is hard to explain a price-earnings ratio of forty-five–just as it is difficult to justify that the land underneath the imperial palace in Tokyo could cost as much as the entire state of California, as was the case at the peak of the real estate bubble in Japan. While our current knowledge does not provide us with the certainty that these situations are bubbles, it does suggest that completely disregarding these indicators is very risky and leads, on average, to bad decision-making. Faced with these aberrations, central bankers would be foolish not to lean against the wind, especially after seeing the costs a bubble’s burst can have on the real economy.
This was after explaining the views of Greenspan and Bernanke.

The lesson Greenspan learned was how politically costly it was to lean against the wind. He dutifully applied this lesson when real estate prices rose. In fact, he elevated the lesson to a principle (the Greenspan Doctrine) that it was not a responsibility of a central banker to try to lean against the formation of potential bubbles. In academia, the staunchest supporter of this approach was a Princeton University macroeconomist little known in the political world at that time: Ben Bernanke! In a 1999 article with fellow economist Mark Gertler, Bernanke analyzed the impact of monetary policy when prices move away from fundamentals. That this contingency was the object of their analysis illustrates how the emt was losing ground. Their conclusion, however, was that the Fed should not intervene, not only because it is difficult to identify the bubbles but also because “our reading of history is that asset price crashes have done sustained damage to the economy only in cases when monetary policy remained unresponsive or actively reinforced deflationary pressures.” Thus, the case against intervention was not based on the idea that the market always gets it right, but on the premise that the costs of these deviations are relatively minor, with respect to the cost of wrong interventions.

In my view, the least bad environment for microeconomic coordination is slow, steady growth in money expenditures. I favor a 3 percent growth path for Final Sales of Domestic Product. Many other "quasi-monetarists" advocate a 5 percent growth path for NGDP. (This is really just "GDP," but to make sure no one confuses it with real GDP, the "N" refers to "nominal.")

These targets for money expenditures are targets for spending on final goods and services--currently-produced consumer goods and services, capital goods, and government goods and services. Both Final Sales and Nominal GDP measure spending on domestic goods and services, and so spending on foreign goods and services are not included (imports,) and foreign spending on domestic goods and services (exports) is included.

How does this relate to bubbles? Other things being equal, a bubble in some domestically produced good or service directly raises total money expenditures on final goods and services. For example, if there is a bubble in single family homes, including "new" ones, and given the trajectory of spending on other goods and services, the increased prices and real volume of production would push Final Sales or NGDP above target. The monetary authority would need to slow money growth to slow the growth of spending on final goods and services. Such a contractionary policy would be likely be associated with higher market interest rates and slower growth in credit. The purpose of the policy, however, is for money spending to return to its targeted growth path.

Then, spending can resume growing at the targeted rate again. If this "pops" the bubble or preempts it, that is an implication of the rule, but not the purpose of the restriction in money growth. Of course, it is likely that spending in other segments of the economy would slow, as well. And it is possible that the bubble would continue. Prices and the real volume of output in the bubble sector (for example, new family homes,) could continue to grow at an unsustainable rate, and spending on other goods and services grow more slowly.

So, consider such a bubble. It develops while money expenditures continues growing on target. For example, suppose firms slow their rate of purchases of capital goods (investment grows more slowly) and instead, households expand their purchases of homes, including new homes. (Residential investment grows more quickly. ) Resources are shifted out of of the production of other capital goods and instead too much goes into the production of new family homes.

Would Zingales argue that the monetary authority respond by slowing the growth of the quantity of money? Under plausible assumptions, this would result in higher interest rates and slower growth in lending. Perhaps the higher interest rates would deter some of those purchasing new homes. That they don't borrow to fund these houses would be a reason credit grows more slowly. Regardless, slowing the growth of the quantity of money relative to the demand to hold that money will slow the growth rate of expenditures on goods and services. Perhaps some of the reduced expenditures will be on housing.

Of course, it is likely that the slow down in money expenditures would slow the growth in the demands for other goods and services too. The higher interest rates and slower growth in credit would involve reduced spending on goods and services other than housing. It would seem plausible that spending on capital goods, already slow, would take a further hit. By assumption, the bubble was pulling resources out of the production of capital goods, presumably seeking unsustainable capital gains on single family homes. While the contractionary policy might reduce the excessive production of single family homes, rather than shift those resources to the production of capital goods which are being under produced, the result instead is that the effort to pop the bubble contract production of those goods as well.

This, of course, is the point that Greenspan, Bernanke, and Gertler were making. Should the monetary authority generate monetary disequilibrium and disrupt the economy on the hope that this will control a possible bubble? They said no. And they were right.

Suppose the bubble in housing bursts. Whatever constellation of perverse expectations that led people to purchase homes they didn't want breaks apart. The demand for housing falls. It drops because the bubble has burst.

What is the least bad policy response? In my view, it is to keep money expenditures growing on target. Since the bubble involved pulling resources out of the rest of the economy to expand production in the bubble sector, when the bubble finally pops, those other sectors need to expand. Growing demands, prices, and profits in those sectors provide the best signal and incentive to generate that redeployment of resources.

The policy of maintaining money expenditures in the face of a burst bubble is sometimes denigrated as the "Greenspan Put." The assumption is that monetary policy will be used to prevent anyone from taking a loss from falling asset prices. Of course, compared to a policy that allows total money expenditures to fall, keeping money expenditures on target likely dampens to decrease in the demand for the bubble sector. Still, it is important to understand that the point isn't to prevent prices from falling in sectors experiencing a bubble. The demand for the good that had the bubble, (like new family homes) can and should grow more slowly or even shrink. The prices of that good can fall, and sharply. But the demand for other goods and services, say, other capital goods, should grow more quickly.

Suppose, however, that the bubble isn't for something like new family homes, but rather the bubble is for stock--say internet stocks. Since the value of claims of ownership in businesses is not a final good or service, this does not directly raise Final Sales or NGDP.

Superficially, money expenditures targeting requires that the monetary authority stand by and allow the bubble to develop. Even if current earnings of internet firms are low and the price earnings ratios balloon to historic highs, as long as the flow of money expenditures on consumer goods and services, capital goods, and government goods continue to grow at the targeting rate, prices of particular assets, or even assets in general freely adjust based upon supply and demand.

However, there are plausible pathways by which the increase in stock prices will indirectly lead to increased spending on consumer and capital goods. As stock prices rise, this directly increases the net worth of those owning the stock. Because this allows those individuals to fund greater future consumption, they will likely reduce current saving and instead expand current consumption. This is the wealth effect on consumer expenditure.

Similarly, those firms that can issue additional shares (or similar ones) may use the opportunity to sell shares to fund purchases of capital goods. The result is expanded expenditure on final goods and services.

Must the monetary authority wait until some measure of money expenditures deviates from target before it can take action? If there were a feedback rule based upon past deviation of money expenditures from target, that would be true.

However, the "quasi-monetarist" approach to policy is to keep expected money expenditures on the targeted growth path. If a speculative bubble in financial assets will cause money expenditures on final goods and services to rise above target, then the monetary authority should "lean against the wind."

However, the focus of such a policy isn't determining whether some asset price is unrealistically high or growing too fast. For example, if people choose to reduce current consumption in order to take advantage of the unsustainable capital gains on technology stocks, and this is expected to free sufficient resources to produce consumer goods for those who cash out and capital goods funded by new stock issues, then it is not the role of the monetary authority to slow money growth and reduce expected money expenditures on final goods and services. And when those expected capital gains turn to capital losses, and those who lose money begin to save to rebuild wealth and investment spending by tech firms grow more slowly, monetary policy should adjust so that expected spending on final goods and services continues to grow on the targeted path.


  1. Maybe you're not the one I should be asking, but would you mind briefly describing why a constant rate of NGDP growth is (theoretically) expected and therefore particularly desirable?

    What I mean is, when I think of myself at home, I have no reason to believe my rate of personal consumption will constantly grow year after year, so I have a hard time understanding why we would expect this of a country. But everyone is talking about NGDP targeting nowadays; perhaps you wouldn't mind helping me understand the appeal.

  2. Long:

    NGDP isn't consumption. It is consumption and saving (and investment) together.

    That your consumption and saving (and tax payments) should add up to your income should certainly be expected.

    While your productivity, and so, your income, might not grow at a constant rate, the average growth in productivity for the economy as a whole is relatively stable.

    This is what generates the growth of real income in the economy. Leaving aside disequilibrium, real consumption plus real investment (and real saving) grow with real income.

    With money expenditure targeting, your income may grow more or less slowly based on your productivity. And your spending and saving will grow more or less slowly with your income.

    Money expenditures in the economy will grow with the average growth rate productivity in the economy. And so, the volume of money expenditures on consumption plus investment (and saving) will grow with the average growth of productivity.

    And, the price level will rise and fall inversely with deviations of productivity from trend. When productivity grows more rapidly than usual, the price level falls, and so there is deflation. When productivity grows more slowly than trend (or even falls) then the price level rises, and there is inflation.

    A plausible alternative would be to have money expenditures grow with productivity. This would keep the price level stable. As a practical matter, this requires keeping track of what is happening to productivity. Further, if there is an error, reversing the error almost certainly causes more trouble than it is worth.

  3. Learning is the most important thing for trading, if we are good at learning then we can really make it big in any field, but if we are not good at learning then we could struggle badly. I am lucky enough to have great support like OctaFX since they have outstanding demo contest like cTrader, it is weekly based with having 400 USD prizes which really helps me work out well and allows me to be successful without any trouble at all.