Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Tuesday, September 8, 2020

Cochrane on why debt matters

National Debt - HISTORY

The stock of national debt is now larger than our annual national income in the United States. Is this something to worry about? Does it matter how big the debt-to-GDP ratio gets? Is there any limit to how large it can grow and, if so, what is it this limit and what factors determine it? A lot of people have been asking these questions lately. John Cochrane is the latest to opine on these questions here: Debt Matters

I'm not even sure where to begin. I suppose we can start with the famous debt clock pictured on the right. Whenever I look at the debt clock, I'm reminded of James Tobin who, in 1949 remarked:   

The peace of mind of a conscientious American must be disturbed every time he is reminded that his government is 250 billion dollars in debt. He must be shocked by the frequent announcement that every newborn baby is burdened, not with a silver spoon, but with a debt of $1700. 

The national debt is now 100 times larger than it was in 1949. Society has somehow managed to hold itself together since then. At the very least, this suggests we need not pay attention to the debt clock. It does not, however, not mean we shouldn't pay attention to managing the debt. Ironically, worrying about the debt is, in a way, what permits us not to worry about it. The time to start worrying is when we and our elected representatives stop worrying about it. According to John, "The notion that debt matters, that spending must be financed sooner or later by taxes on someone, and that those taxes will be economically destructive, has vanished from Washington discourse on both sides of the aisle." That is, it may be time to start worrying. 

I think there's an element of truth to this. For example, while it's true that the Reagan deficits were large, it's also true that there was strong bipartisan support for "doing something about the growing debt." And it wasn't just words. As Justin Fox reminds us, Congress increased taxes seven times between 1982-93. Well, what about Japan? As I explain here, Japan is a poster child for "worrying about the debt." To make a long story short, the debt-to-GDP ratio in Japan has stabilized (pre-Covid, at least), inflation is below target, and the fiscal authority keeps raising the sales tax. Rightly or wrongly, the Japanese "care" about the national debt--the effect of which is to keep fiscal policy "anchored." 

But what exactly is there to fear if fiscal policy becomes "unanchored?" For a country like the United States, it seems clear that outright default will never happen. U.S. Treasury securities (USTs) are too important for global financial markets. A default may very well trigger a global financial meltdown. The only practical option is to continue rolling over the debt, principal and interest (the latter of which is very low these days). Is there a danger of "bond vigilantes" sending the yields on USTs skyward? Not if the Fed stands ready to keep yields low (related post here on yield curve control). And, in any case, even if the Fed raises (or is expected to raise) its policy rate, the U.S. Treasury can just continue to issue the bills necessary to make the scheduled payments. Treasury securities and Federal Reserve reserves are just different forms of interest-bearing money. To put things another way, the national debt need never be paid back--like money, it can be held in private wealth portfolios forever. The only question is on what terms it will be willingly held.

This last point gets to the question of what can be expected to happen if the debt gets too large (say, because the fiscal authority plans to run large primary budget deficits off into the indefinite future). Much will depend on the evolution of the global demand for USTs. If that demand stops growing while fiscal deficits run unabated, surely we can expect the U.S. dollar to weakened and the domestic price-level to rise. The former is likely to contribute to an export boom, which should serve to close the trade deficit (mitigating the adverse consequences of global imbalances). The latter is likely to promote the growth of nominal GDP.  

Needless to say, an export boom and higher NGDP growth don't sound like disaster scenarios, especially in the current economic environment. John seems to worry that whatever happens, it's likely to happen suddenly and without warning. We know Naples is going down (in the manner of Pompeii c. 79AD), we just don't know when. But how does the lava flow correspond to the economic consequences of a debt crisis? (Keep in mind, we're not talking about a country that issues foreign-denominated debt.)

Should we be worried about hyperinflation? Evidently not, as John does not mention it (see also this nice piece by Francis Coppola). But he does mention something about fiscal capacity (the ability of the fiscal authority to exert command over resources). As I explain here, there are limits to how much seigniorage can be extracted in this manner. To put things another way, there are economic limits to how large the debt-to-GDP ratio can get. But reaching this limit simply means that the required tax (whether direct or indirect via inflation) is high--it does not mean disaster. 

John concludes with the following warning: "The closer we are to that limit, the closer we are to a real crisis when we need that fiscal capacity and its no longer there." This is one of those sentences that starts your head off nodding in agreement. But then you think about it for a minute and wonder what type of "real crisis" he has in mind? If it's a financial crisis, the implied positive money-demand shock (flight-to-safety) is likely to increase fiscal capacity, not diminish it. A war perhaps? In these types of emergencies, the nation bands together and governments use other means to gather the resources necessary (e.g., conscription). 

So, to conclude, I'm not saying that John is wrong. It's just not very clear in my mind how he imagines a U.S. debt crisis to unfold exactly. What is missing here is a model. This is odd because one of John's great strengths is model building. And so my conclusion is that it would be very interesting to follow the logic of his argument through the lens of one of his models. Let's see the model, John! 


  1. David,

    Great blog. I almost have no objection to what you say. Yields are policy variable. Higher NGDP growth reduces debt ratio. Japanese have been able to keep unemployment rate to a ridiculously low level - 2.5%. I seldom hear compliant from my Japanese friend, And Yen remains the world's strongest currency at least in the long-run.
    "John's great strengths is model building". I am not so sure... As his fiscal theory of price level is either useless or wrong. As government debt is just money.
    But then it seems that we are increasingly like MMTers......

    1. While I disagree with your assessment of John's models, note that model-building is useful even if the theory is "wrong." The reason is because we can easily see *how* it is wrong. This is very different from the way most prefer to debate -- with vaguely-formed theories floating around in their heads, with assumptions that are either implicit or even unknown.

      Build a model!

  2. David, Thanks for great post.

    I have one question. You mentioned that if we got to the limit of debt to GDP, the required tax would be high. And I think if tax rate is too high, then economy can be chaotic, making every worker unemployed. Isn't it crisis?