Helicopter money: Monetizing the Fed’s perpetual solvency
Money-financed fiscal policy is special precisely because the central bank cannot go bankrupt.
Twice in the past twenty years, severe economic contractions have caused the Fed to cut interest rates to zero in an attempt to support the economy. Given that additional stimulus may be needed once this point is reached, the inability of traditional monetary policy to improve economic outcomes once interest rates have reached zero has prompted fresh thinking about how to add firepower to the Fed’s arsenal. One widely discussed idea, which is as fascinating as it is controversial, is simply printing new money for the government to spend. This tool, colorfully known as helicopter money, is perhaps less revolutionary than it seems at first glance.1 But in certain cases, helicopter money is truly special, and might just be the future of coordinated monetary and fiscal policy.
The key concept underlying helicopter money is that a dollar is an asset to the private sector, but not a meaningful liability to the government.2 Therefore, by creating new dollars, the government can create wealth. This naïve statement comes with a host of qualifiers, but the asset/liability framework is useful for understanding why helicopter money is unique compared to other forms of monetary policy. Moreover, since the term ‘helicopter money’ has been used in both the popular press and the economic literature to refer to a number of distinct proposals, this framework helps specify exactly what we mean by helicopter money.
In this article, helicopter money will be used to refer to the situation in which public spending is financed by a decrease in the net worth of the central bank. In other words, the Fed is creating new reserves to be spent, without receiving an equivalent asset for those reserves. In the most plausible scenario, new reserves would be added to the Treasury’s account, and the Treasury would spend the money on stimulus checks, public infrastructure projects, or a job-guarantee program.3 This definition clarifies why quantitative easing is not helicopter money. Under QE, new reserves are created, but since they are swapped for an equivalent dollar amount of bonds, the Fed’s net worth does not decrease, and no new wealth is transferred to the private sector.4
Similarly, debt monetization is not helicopter money, so long as the purchases are reversible and made at fair market value. While the money from the bonds sold directly to the Fed5 could be used on public spending, the bonds will ultimately have to be repaid through taxes, making this situation similar to traditional debt-financed fiscal stimulus. Of course, if the Treasury sells bonds to the Fed which are not at fair market value (such as charging a positive amount for zero-coupon perpetual bonds) or if the Fed agrees to roll over the bonds indefinitely (or even allows the Treasury to default) then debt monetization can be a pathway to helicopter money. But it is the deterioration of the central bank’s net worth, not the mere creation of new money, that is the defining feature of helicopter money.
Although financing spending by chipping away at the Fed’s net worth is certainly provocative, the tool is not necessarily a useful one under normal circumstances. To see why, consider that when the Treasury spends any newly printed reserves, they will end up back in the banking system as commercial bank reserves (regardless of whether the public saves or spends the money they receive). The Fed pays commercial banks interest on these reserve balances, at a rate comparable to what the Treasury pays on government debt. Since the Fed turns over profits to the Treasury, this interest expense lowers the profit the Treasury will receive. Therefore, under normal circumstances, helicopter money is no free lunch for the government. The plan is just as expensive as financing fiscal stimulus by selling bonds to the public.6
Under abnormal circumstances, however, this is not necessarily the case. The Treasury essentially owns an option in the profit and loss of the Fed. If the Fed makes money, it has to send it to the Treasury. But if the Fed loses money, the Treasury does not have to subsidize the Fed (after all, the Fed can always create reserves to pay its own bills). While paying interest on reserve balances is an expense for the Fed, it is only an expense for the Treasury if the Fed is profitable.7 If the Fed is persistently loss-making, the interest cost of new reserves is not borne by the Treasury.
An easy way for the Fed to generate losses, of course, is through helicopter money. By increasing the Fed’s liabilities without commensurate increases in assets, interest expenses will come to outstrip interest income. Counterintuitively, using helicopter money makes using helicopter money useful. But by routinely booking losses and taking on increasing liabilities, the Fed will quickly become insolvent. As of last year, the Fed’s assets exceeded its liabilities by a mere $60 billion, an amount that would be dwarfed by any meaningful helicopter money program.
One virtue of controlling currency creation, though, is that the Fed is a bank that cannot go bankrupt. Compare the Fed with a commercial bank, like Citigroup. Theoretically, Citi could execute its own version of helicopter money, by crediting each of its 200 million customer accounts with an extra $100. As deposits are liabilities to a bank, this would increase Citi’s liabilities by $20 billion. With a capital base of around $200 billion, this would immediately lower Citi’s net worth by 10%. The higher leverage might spook creditors, which could lead to a run on the bank and render Citi unable to repay its debts. Even if Citi could survive the bank run, shareholders are likely to revolt, and fire the management team giving away money. Profit-maximizing shareholders, regulators, and the threat of bankruptcy all mean that a commercial bank has limits to how much of its net worth it can set ablaze.
For a central bank that issues its own irredeemable currency, though, solvency considerations are operationally irrelevant. The Fed alone determines the quantity of reserves in the banking system, meaning that a bank run is impossible. The Fed’s shareholders exert no control over its operations. By crediting the Treasury’s account with more and more money, the Fed could allow its own capital to go deeply negative, and issue essentially unlimited amounts of helicopter money. Of course, inflation arising from high public spending could cause the Fed to change its path. But since interest-rate policy is independent of reserve quantity, and because helicopter money should only be undertaken when inflation is already too low, high inflation is unlikely to be of immediate concern to policymakers.8
Although the Fed is not meant to be a profit-maximizing institution, embracing loss-making and technical insolvency is likely to result in Congressional testimonies containing awkward questions about the Fed’s balance sheet. The optics would be atrocious. Were helicopter money to be implemented, it is far more likely to be done by somehow neutralizing the interest paid on reserve balances. But the largest barriers to helicopter money would certainly be political. Such close coordination of monetary and fiscal policy could lead to the Fed acquiescing to Congressional spending demands, resulting in fiscal dominance and the death of the Fed’s independence. Still, helicopter money is proof that central bankers have the tools to raise consumption in an economy operating below potential at the zero lower bound. Deflation is a policy choice.
1: The tool is named after a thought experiment proposed by Milton Friedman, who described a helicopter dropping freshly-printed cash onto people.
2: Cash and reserves are nominally liabilities of the Federal Reserve, but they are irredeemable liabilities. Meaningful liabilities grant creditors claims on the cash flow or assets of debtors. Dollars grant no such claims, and are liabilities only in an accounting sense. Contrast this with bank deposits, which, while denominated in dollars, are both private assets and private liabilities.
3: We follow the definition of helicopter money used here, modified for clarity and with the additional qualifier that the net worth of the central bank must decline. We ignore the case of the Fed printing physical cash for use in helicopter money, because that would be logistical insanity and therefore unrealistic. There are versions of helicopter money which do not involve the Treasury, such as individuals being allowed to create special bank accounts with the Fed which the Fed can then credit with money. But for both simplicity and realism (it is almost certain Congress will not give the Fed the authority to do this for the foreseeable future), this will not be discussed.
4: Of course, the hope is that QE does create additional private sector wealth, but the method is indirect. QE is meant to push the economy toward its full potential by incentivizing consumption through lower interest rates. New wealth has not been directly transferred, though, since the public has swapped their existing bonds for an equivalent amount of money.
5: This is illegal anyway since the Fed is required to purchase Treasury securities on the open market, but we are simply clarifying what is and is not helicopter money.
6: See this article by Ben Bernanke which discusses this fact, including mechanisms to neutralize the new interest that needs to be paid. In chapter 13 of his book 21st Century Monetary Policy, Bernanke also discusses the government’s indifference between debt-finance and money-finance due to interest on reserve balances.
7: An important qualifier here is that if the Fed books a loss in a given year, it can only begin remitting profits back to the Treasury once this loss has been surpassed. Suppose in a given year the Fed loses $10, then for the next three years earns $5 in profit per year. The Treasury would only be entitled to the final $5 profit, since the first $10 would go toward paying down the loss on the balance sheet. If the Fed’s losses are an aberration, and the bank will eventually be profitable again, then the Treasury would still incur an expense on new reserves. But if the Fed commits to simply being a loss-making entity with negative capital, with no expectation of future profits, then expenses do not exist for the Treasury in any real sense.
8: The discussion of inflation with respect to helicopter money deserves more space than this article has room for. Notably, one of the few historical instances of actual helicopter money being issued, in Venice during a plague in the 1630s, had to be reversed because of high inflation. Helicopter money is likely to be more inflationary than debt-financed public spending of the same size because 1) monetary financing does not draw an equivalent amount of dollars out of the economy and 2) the public’s marginal propensity to consume the spending should be higher since the public does not have to eventually repay any debt.
First, it should be kept in mind that helicopter money is a tool to be used when the economy is not operating at full potential, and therefore inflation is lower than central banks want. Moderately higher inflation would be an indication that helicopter money is working, because the economy has recovered enough to actually generate inflation, and because higher inflation incentivizes consumption.
Second, because the velocity of money is not stable, a higher money supply does not mathematically lead to a higher price level, despite monetarist wisdom. Again, slightly higher inflation is basically a goal of helicopter money, but evaluating the tool’s effects on the money supply and concluding it will lead to an uncontrollable rise in inflation is not supported by the data.
Third and finally, the Fed does not give up its power to eventually rein in inflation just because it has lowered its net worth by issuing helicopter money. The Fed determines short-term interest rates by settings rates on reserve balances and on its RRP facility. The Fed can never fail to make these interest payments, since it can create reserves. Further, the Fed can control these rates regardless of the quantity of reserves in the system.
Ultimately, inflation is a limiting factor to the extent to which helicopter money can be deployed, but it is also a tool designed to be used when the zero lower bound has been reached, inflation is still too low, and demand is still insufficient to bring the economy back to potential.