Oh Noes, Not The Gilt Death Spiral!

(Where is my &^%*ing fainting couch?)

Ray Dalio’s musings on rates markets are once again doing the rounds on social media. As anyone who has followed popular rates commentary could guess, he was predicting doom, with the U.K. gilt market being the intended victim this time. This article is a series of my standard rants that I hopefully have not repeated in some time (as a “gift” to new readers). They will cover why popular rates commentary is dysfunctional, and why the idea of a “gilt death spiral” shows the serious defects of conventional thinking about interest rates.

As an immediate disclaimer, I have not been following the situation in the U.K. too closely — too many consulting dollars have been flowing my way. However, all evidence seems to suggest that the only actual problem is that the Labour government has is having a hard time putting forth a coherent economic policy, which reflects the problems they face in keeping a united party message in the face of structurally hostile economics coverage. The Conservatives have no such constraint. Liz Truss can blow up the gilt market, and she is still being fêted as a serious world leader in the press.

Why is Popular Rates Coverage Terrible?

If you are interesting in getting useful commentary on the direction of interest rates, you need to go boring street or independent research produced by people who have “rates strategists” on their business cards (not “economist”). Admittedly, as a non-economist who was once paid to produce rates-related forecasts, I am biased. Nevertheless, the rule held up even when I shifted to being a paid consumer of research. I also want to underline that I do not produce rates forecasts now for two very good reasons: I actually am aware of the legal penalties of producing unsolicited research advice for random strangers on the internet, and such writing also gets stale very rapidly, making it useless for a book.

There are four structural problems associated with any other source of rates commentary.

1. Political bias. The business press is pro-free markets, and government debt is associated with governments. So '“government debt = bad” is a baseline view.

2. Useful rates commentary is boring. “Gilt death spiral” gets a lot more readers than “Actually, nothing interesting is going to happen with gilts based on known information.”

3. The business press (and public) does not really care about bond markets, and is happy to assume that anyone who has made a lot of money doing literally anything other than trading duration is an expert on the direction of interest rates. They have a lot of money, so they are experts on bonds, right? Note that this would not be acceptable in media areas that are taken seriously, like sports coverage. Nobody spends a lot of time responding to the Stanley Cup picks of professional football players.

4. Economists believe that since interest rates appear in all their macro models, they are qualified to speak on them without even considering the possibility of examining the time series of bond yields. Conventional economist thinking is fragmented into internally inconsistent models that they pick and choose “lessons” from, and they keep picking the same wrong “lessons” about government debt.

I will only go after the final point in the rest of this article. The first three points are going to be self-evident truths to anyone who has critically thought about popular press coverage of interest rates. (One of the side effects of working in a business that produces investment research, you have to think about investment research as a business, not a source of truth.) The political angle might be disputed, but not questioning whether you are biased because of politics is only a good idea if you have no skin in the game — which is true for pretty much everybody not trading duration as their job. Even asset allocators can be sloppy with their bond calls and do fine — equity volatility dwarfs bond return volatility, so they are mainly judged on their risk asset allocations.

What is the Problem With “High” Debt/GDP Ratios?

If you are government borrowing in a currency you control — a floating currency sovereign — you do not have to worry about involuntary debt default. You might repudiate your debts (like the Republicans who want to solve the “debt crisis” by forcing a default by not raising the debt ceiling), but that is a self-imposed policy. Even if one can comb the books and find some technical factor that could theoretically force a default, the government just needs to change the stupid rule — such rules do not have any “economic laws” behind them (unlike being unable to produce a foreign currency or gold).

The issue with government spending is that “excessive” spending can lead to inflation. The scare quotes around “excessive” are intentional — there are no good quantitative rules determining what levels of spending are going to hit the “inflationary constraints.” (If we did have such quantitative rules, macroeconomics is a solved problem, and all economists studying it in academia and central banks would need to get real jobs.) This ties into “debt levels” in that fiscal deficits mathematically add to government liability levels, modulo various wacky accounting issues that create a wedge between excess cash spending and the deficit. (With apologies to my haters, government money is just another government liability.)

The extension to debt levels is then the following observation: liability service cost is equal to the average interest rate on the liabilities times the stock of liabilities (by the definition of “average interest rate”). To the extent that nominal interest rates do not fall, an increased stock of debt increases the nominal level of interest spending.

Although innumerate people tend to put forth the theory that this will result in a debt/GDP ratio going to infinity, in practice what we see is that nominal GDP levels tend to accelerate due to inflation, and the debt/GDP ratio ends up being capped.

Fiscal Dominance: The Ugly Relation of Mainstream Economics

The conventional response of conventional economists to the idea that raising interest rates can cause higher inflation is to laugh at the person who suggests it. “You buffoon! All the empirical evidence shows that raising interest rates lowers inflation <neglects pointing to any credible quantitative study that actually demonstrates this>!”

However, when they discuss “fiscal dominance,” they agree exactly with the premise of the previous section — which shows that higher interest rates might lead to inflation. Since this disagrees with the dogma in the previous paragraph, the discussion of “fiscal dominance” is confined to “fiscal dominance is bad, and should not happen.”

The apparent exception to this incoherence is the Fiscal Theory of the Price Level. That theory makes the following bold quantitative prediction: the price level instantaneously jumps in response to announced fiscal policy changes. The reality that this prediction is quite obviously not supported by any data has not dampened the enthusiasm of believers in this theory. Although that outcome would be surprising for anyone who studied a real applied science in university, anthropologists would have no difficulty in explaining it.

Economics 101-ism

The conventional way to sidestep the previous problem is to rely on Economics 101, in particular supply-demand curve arguments. (Supply and demand curves are the last refuge of the innumerate.)

The argument is straightforward: if we increase the supply of government bonds (i.e., the stock outstanding) and hold all else equal, their price will fall (yields go up). It’s science, amirite?

Even if we granted the premise of supply and demand curves for government debt, the “all else equal” assumption fails. Government deficits inject money — a government liability — into the hands of the non-government sector. If you bizarrely take supply/demand curves seriously, a fiscal deficit moves both the supply and demand curves. Central government deficits for currency sovereigns are self-financing. There are practical limits to how large a given auction can be for it not to be a fiasco, but those limits scale with the size of the pre-existing financial system. That is, there is a practical upper limit for nominal debt growth, that upper limit is far higher than any plausible need for a government that is not running an experiment trying to break the financial system.

Modern Monetary Theory (MMT) gained a decent following among bond traders and analysts because the MMT proponents dropped the Economics 101 nonsense and just said that bond yields are the relative price between cash and long duration instruments, and that they are best understood as mainly determined by the expected path of the policy rate — which is set by a committee of pointy-headed economists, not “markets.” The interesting thing about this observation is that neoclassical models are also built around this premise (as are fixed income pricing models!), but for some reason, all the believers in neoclassical theory for some reason use the previous “Economics 101” story when writing rates commentary. Go figure.

Concluding Remarks

Popular rates commentary is to be consumed as a source of entertainment, not enlightenment.