Saturday, November 13, 2010

Why So Many Critics of QE2?

Why are many free market economists so critical of the Fed's proposed quantitative easing? I will suggest four reasons: the quantity theory of money, the public finance view of money creation, a focus on growth rates rather than levels, and an assumption of market clearing.

The first "problem" is the quantity theory of money. The quantity theory of money provides many valuable insights, but sometimes can be misleading. My bottom line version of the quantity theory of money is simple. Given the nominal quantity of money and the real demand to hold money, the price level adjusts so that the real quantity of money accommodates the real demand to hold money.

The Fed proposes to purchase $600 billion in government bonds. The monetary base is approximately $2 trillion, so the Fed is proposing a 30 percent increase in the nominal quantity of money in less than a year. Ceteris paribus, the price level must rise 30 percent so that the real quantity of money adjusts to this higher nominal quantity of money. And so, the quantity theory of money suggests that quantitative easing is highly inflationary.

The fundamental problem with this approach is that if a monetary authority is constrained by a rule controlling any nominal quantity other than base money itself--the inflation rate, the price level, nominal income, or even the price of gold, then the quantity of base money can't be treated as given. (The quantity of base money can't be taken as given with a rule controlling the nominal interest rate, but that sort of target is so counterproductive it is not worth considering.)

The Fed appears to be targeting a 2 percent inflation rate, so increasing base money by 30 percent and then waiting for some kind of long run impact on the price level (particularly a 30 percent increase) violates the rule. If the resulting inflation rate is anything more than 2 percent, then the Fed will be bound by its rule to reverse course and decrease base money so that no more than an additional 2 percent inflation occurs.

If the Fed had an explicit price level rule or a rule for money expenditures, the tentative nature of any increase in base money would be clearer. If an increase in base money, however large, causes the price level or money expenditures to rise above its target value, then it must be reversed. Whatever the long run effect of a given $2.6 trillion of base money might be for the price level, the actual quantity of base money will change the amount necessary to hit the Fed's nominal target.

The second problem is the public finance view of money creation. From this perspective, government can finance spending by collecting taxes, borrowing, or issuing new money and spending it. Especially when combined with the quantity theory of money, the implication seems to be that the desire of government to spend determines the quantity of money and the quantity of money then determines the price level. From this perspective, changes in the quantity of base money are one way--every increase is permanent, having funded new government spending. In particular, the 30 percent increase in the quantity of base money is funding part of the recent increases in government spending, and so, should eventually result in a 30 percent increase in the price level.

This approach applies to an unconstrained monetary authority-- the Zimbabwe of recent times, for example. But if a monetary authority is constrained by any nominal target other than the quantity of base money itself, then money creation is better understood as borrowing. The monetary authority can issue zero interest hand-to-hand currency, borrowing at a zero nominal rate. Reserve balances are much the same, with the monetary authority paying the interest rate it chooses on the balances banks keep with it.

The reason the issue of base money is best understood as borrowing is that the monetary authority must stand ready to reduce the quantity of base money if necessary to meet the target. While the government might create money to spend, it must stand ready to issue more conventional, interest bearing debt, to withdraw that money from circulation if needed to meet its target.

The Fed operates as an independent central bank. It borrows by issuing hand-to-hand currency and reserves which allows it to fund a portfolio of assets. Traditionally, it mostly held government bonds of a variety of maturities, but recently, it went through a period where it made a large amount of loans and currently holds a substantial portfolio of mortgage backed securities along with government bonds. Quantitative easing involves the Fed borrowing $600 billion, most probably from banks that will hold reserve balances paying .25 percent interest and then lending the money by purchasing $600 billion of government bonds. As already explained, if this causes inflation to rise beyond the Fed's 2 percent target, the Fed will need to sell off some of its assets, perhaps these bonds it has purchased, and "pay off" the currency or reserves it issued to fund them.


With a gold standard, the obligation to redeem the paper money issued by the monetary authority--currency and reserves-- with gold shows that it is a type of debt. If the demand for this type of debt should rise substantially, then a large increase in the quantity is possible. The monetary authority can borrow at a zero interest rate (or at whatever interest rate it chooses to pay banks on their reserve balances.) However, if the demand for currency and reserves should fall, then the monetary authority must pay it back. For the Fed, "paying it back" would involve selling off securities in its asset portfolio. Of course, if the Fed refused to contract the quantity of base money to match the decreased demand, redemptions of currency and reserves for gold would force it to do so.

While a rule for inflation, the price level, or money expenditures would not have such a simple enforcement mechanism, if the rule is somehow enforced, the Fed would similarly be obligated to reduce the quantity of base money--currency and reserves--to match any decrease in demand. With such a rule, that currency and reserves show up as liabilities on the Fed's balance sheet is no illusion. And, the government cannot simply print money and spend it. It can run a budget deficit and finance the national debt by borrowing through the issue of currency and reserves--but only to the limit that follows from the amount households and firms are willing to lend given the rule.

A third problem is excessive focus on growth rates rather than levels. Since increasing amounts of labor and capital along with improving technology results in a growing productive capacity for the economy, growth rates are important. During the Great Moderation, a growing quantity of base money generated a growing quantity of currency and bank deposits available for households and firms to hold, money expenditures growing at an annual rate of about 5 percent, prices rising about 2 percent per year, and real expenditures growing about 3 percent per year, which matched the approximate 3 percent growth rate of productive capacity. Employment grew about 2 percent a year, which matched the 2 percent growth rate of the labor force.

For the economy to remain in equilibrium, all of these growth rates need to balance. If the economy starts in equilibrium, then as long as the growth rates are appropriate, the economy will stay in equilibrium. If the economy starts in equilibrium, and the growth rate of base money shifts to 30 percent, the eventual result would be that money expenditures would grow approximately 30 percent per year, and the inflation rate would rise from 2 percent to 27 percent. Quantitative easing looks bad.

Further, an examination of recent growth rates suggest that the economy isn't far from equilibrium. The growth of money expenditures as measured by Final Sales of Domestic Product in the third quarter of 2010 was 2.8 percent. The growth rate of real output was approximately 2.4 percent, and the inflation rate was 2.25 percent. (Nominal GDP grew faster than Final Sales of Domestic Product because of a substantial build up of inventories.)

While the growth rate of real output of 2.4 percent is a bit low, it is higher than the CBO estimate of the growth rate of the productive capacity of the economy--1.5 percent. Personally, I consider a 2.8 percent growth rate for money expenditures to be near ideal. And not only is the inflation rate of 2.25 percent above the Fed's vague target of 2 percent, my view is that zero inflation is the proper goal. How do these figures come close to justifying a 30 percent growth rate for base money? Supply-side reforms aimed at getting productivity to rise looks to be the proper policy response.

But this is all an illusion created by excessive focus on growth rates. If the economy were at equilibrium, then that would be reasonable enough, but money expenditures are nearly 13 percent below the level of the Great Moderation. The price level is 2 percent below the level of the Great Moderation. Real GDP is about 6.4 percent below the CBO estimate of potential income and 10 percent below growth path of the Great Moderation.

The more than 100 percent increase in base money over the past two years--about $1 trillion-- has had only a modest impact on the measures of the quantity of money available for households and firms to hold. A one time, mostly likely temporary, increase in the monetary base of 30 percent is not out of line with the large apparent disequilibrium in growth paths. In particular, returning money expenditures to the growth path of the Great Moderation would require a 21 percent increase over the next year. It is very possible that the increase in the monetary base would be too little.

The final problem is the principle of market clearing. The concept of market clearing is an essential building block of economic analysis. The notion that any surplus--desired sales greater than desired purchases--will promptly result in a price sufficiently low so that the plans of the sellers and buyers match has important macroeconomic implications. For example, other things being equal, if money expenditures fall to a growth path 13 percent below the Great Moderation, the price level should simply fall in proportion--13 percent below its growth path. Otherwise there would be surpluses of various goods and services, implying that the price level is too high. Once the price level falls 13 percent, the flow of real expenditures rises back to its previous growth path, equal to the unchanged growth path of the productive capacity of the economy.

The price level, however, has only fallen 2 percent below its growth path of the Great Moderation and real expenditure is about 11 percent below its growth path. Assuming the growth path of the productive capacity is unchanged, there should be massive surpluses of goods and services. While firms would like to sell more, they cannot, and so reduce production to match the depressed real volume of sales. Needing less labor to produce this lower level of output, hiring should fall, perhaps layoffs expand. The surpluses of goods turn into surpluses of labor. When asked why they are not producing more and hiring more, firms might be expected to respond that their problem is weak sales. When asked why they are not working, the unemployed might complain that there are not enough job openings.

But this scenario is inconsistent with the principle of market clearing. If there were surpluses of goods and resources, then the price level would fall. If the price level has only fallen 2 percent, it must be because the existing level of real expenditures (approximately 11 percent below the growth path of the Great Moderation,) is equal to the current productive capacity of the economy. The principle of market clearing implies that the productive capacity of the economy must be 11 percent below that of the Great Moderation. While the CBO estimate suggests that the productive capacity of the economy has only fallen 4 percent below the growth path of the Great Moderation, their estimation procedure is inconsistent with the principle of market clearing and must somehow be in error.

If polling shows that firms claim they are not producing more because of weak sales, this must be an illusion. The principle of market clearing clearing implies that the problem must be bottlenecks of key resources. Workers may complain about an inability to find jobs, but the principle of market implies that problem must be that workers would prefer to collect extended unemployment benefits, or are choosing to retire early because of high taxes, or that the unemployment is limited to unskilled workers constrained by the minimum wage from working at lower pay.

If the productive capacity of the economy really has fallen about 11 percent, and the natural rate of unemployment has risen to about 9.4 percent, then the economy is in equilibrium, and the current growth rates in money expenditures, inflation, and real output are roughly in balance. I doubt it.

I think the most reasonable assessment of the economy is that prices and wages remain "too high" for the current growth path of money expenditures, and that rather than waiting for what appears to be a very slow process of market adjustment through lower prices and wages--perhaps complicated by expectations that a recovery of money expenditures is eminent--the better approach is to reverse the decrease in money expenditures and get them back to a reasonable growth path.

I favor a modified growth path for money expenditures--3 percent rather than 5 percent. The Fed should adjust the quantity of base money, increase it or decrease it, whatever amount is needed to stay on that growth path. In effect, the Fed can and should borrow by issuing currency and reserves in a variable amount. If the demand for base money rises, so that banks, other firms, and households want to lend a large amount to the Fed, then the Fed should borrow that amount. And if the demand for base money falls, then the Fed needs to pay it all back by selling off some of the assets it has accumulated. If the Fed stays on target, then the growth rates should all stay in balance, but if it fails, then the growth rates should adjust to return to the target growth path. Finally, if the productive capacity of the economy changes--grows more rapidly or more slowly, or even falls, then the growth path of the price level, and so the inflation rate, should be allowed to change.

From my perspective, the Fed allowed money expenditures to fall and money expenditures are far too low. Increasing base money through quantitative easing is a move in the right direction. Treating the increase in base money as given, or as funding government spending, or as a permanent increase in the growth rate are mistaken. Looking at current growth rates at very low levels is a mistake. As is treating market clearing as a principle, and failing to recognize an economy suffering from a depressed level of real expenditures.

1 comment:

  1. Quantity theory? Nah, I'm just terrified of what commodities are doing right now.

    ReplyDelete