The New Republic, August 1, 2011
Economists believe that people respond to incentives. The fact that economists never suffer career consequences for failing to consider new ideas explains why they so rarely consider any policy that has not long been in the standard bag of tricks. I mention this background since it is relevant to the reaction given a proposal on the debt ceiling that Ron Paul originally put forward and that I subsequently endorsed. Paul suggested that the Fed could destroy the $1.6 trillion in government bonds that it now holds as a way of getting room under the debt ceiling. Debt to the Fed counts as part of the government debt subject to the limit. If the Fed destroyed $1.6 trillion in debt, then it would create a space of $1.6 trillion under the ceiling.
This is an interesting way of getting around the ceiling, although it would almost certainly require an act of Congress to do it. As it turns out, the other side of this story is even more interesting. The Fed plans to sell off the $1.6 trillion in government bonds it currently holds. It also plans to sell off more than $1 trillion in mortgage-backed securities it bought to help stabilize financial markets at the peak of the financial crisis. Following the logic of Paul's idea, I suggest that the Fed could simply hold on to large amounts of debt for an indefinite period of time. The interest on this debt would continue to be paid to the Fed and then be refunded to the Treasury—an effective and easy way to reduce the deficit that almost no one is talking about.
As long as the Fed holds onto the bonds that it currently holds, it receives the interest on them. Last year, the Fed refunded almost $80 billion in interest to the Treasury. Once the Fed sells off its assets, it will no longer be issuing these large refunds. Instead, the interest on the Treasury bonds currently held by the Fed will be paid out to the general public and will impose a burden on the federal budget.
If the Fed holds onto the bonds, by contrast, the potential savings to the Treasury are substantial. By holding onto its $3 trillion in debt (roughly its current level) over the next decade, it will earn an income of close to $120 billion a year. Taking this sum over the next ten years and adding interest savings on the debt would push the total savings to the government to almost $1.4 trillion. In other words, this is real money.
There is a downside to going this route. If the Fed holds on to these bonds, it would imply a huge expansion in the money supply, since the Fed issued trillions of dollars in new reserves to buy bonds over the last three years. This huge amount of new reserves could lead to inflation once the economy recovers. The Fed had planned to sell its bonds to pull the reserves out of the system; if it holds them to gain the interest instead, it cannot sell them to reduce reserves.
An answer to this problem is to raise the reserve requirement on bank deposits. If the amount of reserves is doubled, but the reserve requirement is doubled also, there should be no inflationary impact from the higher level of reserves. Higher bank reserve requirements do present a greater opportunity for the shadow banking system to get around this regulation. For example, if banks have to hold 20 percent of their deposits on reserve, they could be undercut by brokerage houses or other financial institutions that are not subject to the same reserve requirement. Fortunately, the new Dodd-Frank financial legislation gave the Fed and other regulators increased authority over the non-bank portions of the financial system, which means that they should be better positioned to ensure that shadow banks do not simply undermine the higher reserve requirements imposed on banks.
Of course, it is not possible to know in advance if it will be possible to rely on reserve requirements to choke off whatever inflationary pressures might develop. But it seems worth a try. After all, how many voters would prefer the $1.4 trillion in cuts to Social Security, Medicare and Medicaid, which was on the table as part of a grand deficit reduction deal?
Besides, if efforts to contain inflation through higher reserve requirements proved ineffective, it would always be possible to go back to the Fed's plan of selling off the reserves. There are no economic models that show inflation soaring into the double digits overnight. We would have time to see any problems with inflation developing and then to respond accordingly.
At the end of the day, perhaps there are good reasons for not wanting to try this experiment, but I have yet to see any. Given that $1.4 trillion is real money, even in Washington, it might be worth a few minutes of thought from economists in policy positions. But, as we know, no economists will lose their job or even miss a promotion if we end up cutting Medicaid for poor children or throw seniors out of nursing homes. And economic theory tells us if you don't give people incentive to be creative thinkers, then they won't be creative thinkers. We have exactly the sort of people designing economic policy that economic theory predicts.