Sunday, October 2, 2011

Helicopter Money--What Does it Mean?

Helicopter money has been getting more coverage. Matt Yglesias likes the idea.

Old monetarists, including Milton Friedman, sometimes would use the metaphor of helicopter money. The monetary authority would load up newly-printed, hand-to-hand currency into helicopters and then fly around throwing it out. People would catch the falling money or else rake it up from the ground. The idea is that they would spend the money, so that an increase in the quantity of money dropped from helicopters would generate additional spending on current output.

If there were some problem with inadequate aggregate demand, the helicopters could fix it. Supposedly, firms have reduced output and employment because sales are low. Once spending picks up, they produce more, sell more, and hire more employees.

Of course, if inadequate demand were not the problem, then nominal expenditure would still rise, and the result would simply be higher prices, including higher resources prices like wages. Excessive money creation would lead to inflation.

Implicit in the helicopter story is that notion that fiat currency is like paper gold. But rather than costly digging, the government just prints it up. Just like gold miners sell, or spend, their product without any thought about what this does to the purchasing power of gold, the government spends the paper money it prints without worrying about the price level.

Old monetarists, of course, strongly opposed having an unconstrained monetary authority. However, their answer was to put some kind of constraint on the quantity of money. A fixed quantity of money would be one possibility, but generally they are better known for advocating slow, steady growth in the money supply. This is best understood as having some measure of the quantity of money on a slow, steady growth path.

While the old monetarists expected this monetary regime to result in a more or less stable price level, from the point of view of the government creating the money, it is just printing it and spending it. There is no worry about the price level or anything else, just an artificial limit on how much money it can obtain for "free." In other words, rather than the government obtaining seigniorage revenue without constraint, there are limits.

Allowing the government no free money, freezing the monetary base, results in a deflation rate equal to the growth rate of real output. Assuming the equilibrium real interest rate is equal to the growth rate of real output, then the real rate of return on currency would be equal to the real interest rate. In this scenario, the real quantity of currency held would rise (and the price level fall) until it provides liquidity services on the margin equal to the near-zero cost of printing the currency. This maximizes the total liquidity services provided by the money.

However, this "optimal" regime was usually ignored in favor of a growth rate for the quantity of money matching the trend growth rate of real output. Slightly slower rates of money growth generates less seignorage for the government and modest deflation, moving closer to the optimal regime described above. Faster rates of money growth, however, results in higher inflation rates, and more revenue for the government. Households and firms pay a higher inflation tax, the amount of money held is less optimal (like any other tax base,) and the government collects more tax revenue.

Considering a faster rate of money growth, how does this impact the government's budget? Using the framing described above, the government has additional revenue. It is collecting more from the "inflation tax."

How might it respond? While it could use the added funds to purchase goods and services (hopefully public goods of some sort,) it might also reduce other taxes. If the government does cut other taxes, those who were paying those taxes, have received a "helicopter drop" of money. Or, perhaps the money could be spent by expanding transfers. For example, payments could be made to those who are unemployed. Again, this is like a helicopter drop of money.

However, an alternative framing is that the difference between government expenditure and receipts is a budget deficit, and money creation is a method of funding the budget deficit. The alternative method is borrowing, generally, by selling interest-bearing government bonds.

With this framing, given the government budget deficit, more rapid money growth directly reduces the amount the government borrows by selling bonds. Suppose the government is running a budget deficit funded partly by selling bonds and partially by issuing money. More rapid money growth reduces the part of the budget deficit funded by borrowing, and has no direct impact on taxes, transfers, or government spending.

There is no helicopter drop. Ceteris paribus, the government borrows less, and given the supply of loans, more funds are available for everyone else to borrow. Basic supply and demand analysis suggests that the surplus of funds results in lower interest rates, and an increase in the quantity of credit demanded. Firms borrow to fund capital expenditures. Households borrow to fund consumer expenditures, particularly consumer durables. Suddenly, money creation is operating through a credit channel.

A lower interest rate should also reduce the quantity of credit supplied. The government is borrowing less, leading to a decrease in the demand for credit, and the resulting decrease in the interest rate makes lending less attractive. To the degree that those who reduce lending use the funds to purchase capital goods or consumer goods, then the effect is qualitatively no different from an increase in the quantity of credit demanded.

However, it would also be possible for those who reduce their lending to simply hold money. And so, the more rapid growth in supply of money might have the peculiar effect of causing more rapid growth in the demand to hold money. If this effect would sufficiently extreme, to the point where the increase in money growth is matched one-for-one with additional money demand, the result is the so-called "the liquidity trap."

Suppose the money creation reaches a point where it is equal to the government budget deficit. The government is borrowing nothing by selling bonds. Still, the effect is a decrease in the demand for credit by the government, a surplus of credit, lower interest rates, and more spending by firms as households as they lend less or borrow more.

Suppose money creation is greater than the government's budget deficit? This appears to create a paradox with the "inflation tax" framing, but it is certainly possible. The government creates new money and uses the proceeds to pay down the existing national debt. Again, the effect is reduced government borrowing, lower interest rates, and so, additional spending as firms and households reduced the quantity of credit supplied or raise the quantity of credit demanded.

Suppose, however, that rather than hold the government budget deficit constant, taxes are cut, transfer payments are increased, or government expands spending on goods and services. In this scenario, particularly with the tax cuts and increased transfers, there is a helicopter drop of new money. On this framing, the "helicopter drop," is fiscal policy. Tax cuts or additional government outlays, expand the government's budget deficit.

And, of course, this is exactly what has happened in the U.S. over the past three years. Taxes have been cut and transfer payments have expanded, and there have been increase in federal government spending. The budget deficit has grown tremendously, along with the more rapid growth in the quantity of money.

The helicopters have been running!

The monetary base has increased by approximately $1.6 trillion. The total federal debt has increased by $4 trillion. Of course, the federal government was already running budget deficits in 2007, but that doubled in 2008, to over $400 billion. For 2009, it was $1.4 trillion. And then in 2010, $1.3 trillion. Treating the last three years as if the government was simply funding more of a given budget deficit out of money creation is far removed from reality. Consolidating the Federal Reserve and Treasury, the government has created large amounts of money and spent it, one way or another.

However, there is a deeper problem with the old monetarist framing. The problem with the approach is not simply that the revenues created by the inflation tax aren't always spent or used to reduce other taxes, and so can simply impact government borrowing. It is rather that the "paper gold" approach to fiat money is flawed. It does apply to a truly irresponsible government that just creates money and spends it without any concern about its purchasing power. It also applies to a government that is subject to a quantitative limit to its money creation.

However, suppose that the monetary authority is committed to some other kind of nominal target? (For market monetarists, this is the normatively relevant scenario, because we favor a nominal GDP level targeting. Or, as I always say, a target for the growth path of nominal GDP.) This would apply to a monetary authority targeting a growth path for the price level, or even for the inflation rate. If this is true, then the government is not simply printing money and spending it, with or without some quantitative restraint.

Under this scenario, when the government "prints money," (issues more hand-to-hand currency,) it is borrowing. If it is issuing currency, it pays no interest. Also, there is no way that outsiders can demand "repayment" of those loans. While those who hold the currency are making a loan to the government, if they don't want to lend in that fashion any longer, all they can do is spend the currency on some other good, service, or asset. And that includes depositing money in a bank, and substituting currency for deposits. However, if the demand to hold currency falls, the monetary authority must reduce the quantity. It must pay those loans back.

If the monetary authority is consolidated with the government, as the "old monetarist" scenario assumes, then some or all of the national debt is being funded at zero interest by issuing currency. And if the demand to hold currency falls, and the monetary authority is to keep to its nominal target (say for inflation,) then more of the national debt must be funded by issuing interest-bearing debt. (Of course, it would also be possible to raise taxes or cut government spending when the demand for currency falls, reducing the budget deficit to reduce currency issue.)

Again, assuming a commitment to a nominal target, like a growth path of nominal GDP or inflation, even if the monetary authority were to print up currency and load it into a helicopter and push in out so that people could catch it as it falls or pick it up off the ground, it would still be borrowing money. If this is a new type of transfer program, and other taxes and government spending don't change, it is a higher budget deficit. The national debt increases.

While it is funded in the first instance by an issue of hand-to-hand currency, and has no immediate interest expense, in the long run, it implies either higher tax payments or reductions in future government spending. Unless a remarkably high demand for base money is expected to be permanent, this exacerbates the already unsound long run financial condition of the U.S. federal government.

As for its effectiveness in expanding demand, it is no different from the tax cuts and increased transfers funded in the first instance by government bonds, but combined with ordinary open market operations by the Fed, purchasing government bonds with newly issued base money. There is really nothing special about helicopter drops of currency.

If the goal is to expand nominal expenditures while avoiding further expansions in federal budget deficits, or better yet, reducing them, conventional open market operations are necessary. For a monetary authority operating on banking principles, like the Fed, it must purchase assets with nearly created money, not print up money and give it away. From the point of view of the entire government, including the Fed, more of the national debt must be financed by money, rather than interest bearing debt.

Of course, if people became convinced that there has been a regime change where money is being created and spent without any concern about its future purchasing power, the result could be highly inflationary, so that any concerns about inadequate demand would rapidly disappear. Those who suggest that the Fed take lessons from Zimbabwe are suggesting such a change. If Bernanke begins to give Obama money to pay off supporters, and this was done openly, inflation should be no problem.

Further, if the new regime were simply that the monetary base would be fixed or grow slowly from its current level of $2.6 trillion, the result would also likely be highly inflationary. Those hard money advocates who treat freezing the monetary base as an ideal, or as a second best option, should take care about where the U.S. stands today.

And what about a third approach? Suppose the new regime was simply a target for the growth path of nominal GDP? Sure, hand-to-hand currency would still be a type of debt. With a given budget deficit, or a smaller one, no one is receiving new money as a gift that can be spent as if it was falling from the sky.

With a central bank operating on banking principles, like the Fed, it is purchasing financial assets. Consolidated with the rest of the government, less interest bearing government debt is being issued. Ceteris paribus, the effect would be higher prices for government bonds of some sort and lower interest rates. To the degree lower interest rates provide a signal and incentive for firms and households to purchase more capital and consumer goods and services, demand increases.

But the expectation that the Fed will purchase enough government bonds (or whatever other assets are needed) to reach that goal, or alternatively, the government as a whole will fund as much of the national debt with currency as needed to reach that goal, then expectations of higher nominal GDP can result in higher, rather than lower interest rates. It simply requires that firms and households sell more bonds than the Fed buys. And even more paradoxically, it is possible that expectations that nominal GDP will rise to target will result in such a decrease in the demand to hold base money, that the Fed would need to shrink the quantity of money.

There is no need for helicopter drops of money. A monetary regime that can be usefully understood with the helicopter drop metaphor is undesirable. What is needed instead is a clear target for the monetary authority--a target for the growth path of nominal GDP--and a willingness to purchase financial assets at market prices in whatever quantities are necessary to hit the target.


















No comments:

Post a Comment