Does it matter that we can print money?

As far as the budget is concerned, not really.

According to Randall Wray, government default is never really a concern, because “a sovereign government can always make payments as they come due by crediting bank accounts”. This is true in a trivial sense: as long as the government defines what a “dollar” is and can create it at zero cost, it can always make payments on its nominal debt. But this is virtually irrelevant as a matter of actual policy—although the government could theoretical print all the money needed to fulfill its obligations, the outcome would be far, far worse than simple default.

To see why, let’s distinguish between two ways that control of monetary policy can ease the debt burden: seignorage and inflation. Seignorage is the return from creating money that pays below the prevailing rate of interest—usually zero percent, though not always. By financing part of its debt through currency that pays less than the rate on T-Bills, the feds can manage their debt a little more cheaply. But this is only a very marginal part of the budget: under normal circumstances, the monetary base would be around $1 trillion, and the interest rate on 3-month T-bills might be 4%. This adds up to $40 billion a year—a respectable sum, to be sure, but not much next to a deficit of $1.7 trillion.

Couldn’t the government increase its revenue from seignorage? Perhaps, but not by very much, and the effects would be swamped by the other macroeconomic factors in play. The Fed can’t directly control how many dollars the world is willing to hold: it can only adjust the nominal interest rate, changing the cost of holding wealth in the form of cash. A higher interest rate means more seignorage revenue per dollar in circulation, but it also makes the public demand less cash—at the most basic level, this is no different from any other way of funding the government. We even have the Laffer curve: once the long-term interest rate is high enough, further increases will depress demand for cash so much that the increase in seignorage revenue per dollar will be offset by the decline in dollars held by the public. This is particularly an issue for the US, where most of the $1 trillion in paper currency is held as a store of value (much of it outside the country) rather than a medium of exchange. If the Fed decided to implement a higher long-term interest rate (and therefore a higher long-term rate of inflation), it wouldn’t take long for the drug lords of the world to find an alternative.

(But wait—the Fed prints dollars, right? If it decides to print another trillion, who cares what the public thinks—won’t they be forced to hold the money and pay the low rate, like it or not? Not really. If the Fed prints more money than the public is willing to hold at the current interest rate, something has to happen for supply and demand to equate, and that “something” is inflation. Of course, since prices are nearly fixed in the short term, in practice what happens when the Fed changes the money supply is that interest rates move; if the Fed creates $1 trillion in reserve deposits, the interest rate would plummet to 0%. But this is hardly a good way of earning more from seignorage: at 0%, you’re earning nothing at all! Even the Fed can’t escape a demand curve.)

The conclusion is inevitable: seignorage can play only a very, very small role in resolving our fiscal problems.

There is, of course, another way that monetary policy can affect the debt: inflation. Except for the small fraction held in the form of TIPS, the Treasury’s debt is all nominal. By engineering a surprise inflation, the Federal Reserve could lower the real value of the debt, making it easier to pay back in the future. But there are several problems…

First, Treasury debt has a very short average maturity, clocking in at about 5 years. As soon as the “inflate away the debt” policy became clear, investors would demand far higher nominal interest rates as a precaution—or, more likely, they’d shy away from nominal dollar debts altogether, demanding either inflation-protected debt or debt in another currency. The only fiscal benefits from inflation, then, come from debt that’s already outstanding—and the short average maturity means that inflation doesn’t have much time to eat away at that debt. If we increased trend inflation by 8% (to, say, 10%), our real debt burden might fall by 40%—at the cost of crippling the ability to borrow affordably in our own currency, not to mention the macroeconomic fallout from a reckless monetary authority. And the real fiscal problem isn’t the current debt: it’s the implicit liability from entitlement programs, a liability that inflation can’t erase.

Second, and just as importantly, a conscious policy of inflating away the debt is really just an implicit default. After experiencing such a policy, certainly investors wouldn’t trust nominal dollar debts—but it’s also possible that they would be skeptical of American credit in general. After all, if the US cheats on its debt once, it’s not clear why investors should expect anything different in the future: maybe the specific tool of eroding the value of nominal debt wouldn’t be available, but the feds could find another way to skate on their obligations (perhaps an explicit default). Only in a very strange model of credibility would investors say “well, you deviated from your commitments by doing X, but now that you no longer can do X we have absolute faith in your commitments going forward”. And that would be an extraordinarily, extraordinarily costly move: the US can currently borrow cheaply because its debt is viewed as a safe, liquid store of value by investors the world over. The long-term increase in financing costs would almost surely swamp the short-term decline in debt.

And that brings us back to Wray’s claim: that the Fed’s ability to create money makes default impossible. As I mentioned, seignorage revenue can’t possibly approach the levels necessary to tackle the deficit—and as a matter of arithmetic, this means that the monetary option for avoiding default would have to work almost exclusively by inflating away the debt. Since the maturity of American debt is so short, this would only be possible through an immediate and extraordinary increase in inflation, to levels that would make the 70s look like a picnic. And all this would merely be a roundabout way of defaulting on the debt—but one with even more overwhelming macroeconomic consequences than an explicit default!

So yes, the Fed’s ability to create money technically means that the US can repay its debts. But only in the most practically useless sense imaginable.

1 Comment

Filed under macro

One response to “Does it matter that we can print money?

  1. Pingback: Taxes, hidden and explicit | Matt Rognlie

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s