Bond Bear Market Scare Stories

Whenever there is an uptick in bond yields, scare stories about a coming secular bond bear market are not far behind. The problem with most of these stories is that they are not particularly compelling, and different people have been invoking variations of them for decades. Instead, if you want to come up with a much scarier bond bear market scenario, we need to drop some analytical assumptions, and think through the implications.

The Lame Scare Stories

Each commentator comes up with a different spin on why the Treasury market is about to collapse, and what the implications are. I cannot hope to cover them all. However, there are a few basic stories that quite often appear. Unfortunately, if we want to translate them to a (highly dated) pop cultural reference, they are about as scary as Dr. Tongue's Evil House of Pancakes.

Since I am only summarily dismissing these arguments, I will not waste the reader's time by trying to relate them to particular analyses.

The first (and most common) scare story is about rising inflation. (Admittedly, I always throw in a disclaimer about this scenario when discussing long-term prospects.) The problem is that we almost have three decades of stable inflation (since the early 1990s) in the developed countries. This period also included two oil price spikes, which did not translate into higher inflation. (In fact, they preceded recessions that led to lower inflation.) Meanwhile, commentators have been calling for an inflationary accident throughout that entire period. It is clear that we need some form of structural change to help sustain higher inflation.

The second scare story revolves around foreign central banks suddenly dumping Treasurys. These stories fall apart when we realise that even foreign central banks do not want to vapourise their capital. Furthermore, one needs to explain how the flows that lead to this liquidation will be sustained. Selling Treasurys implies a falling U.S. dollar -- making other countries exporters less competitive. Why exactly do these central banks want to sabotage their existing trade policy? Although it is possible to think of scenarios justifying such an outcome, there are a lot of moving parts in the stories that can break down.

The last set of stories are in the "who will buy the bonds?" category. Since monetary flows are circular, these stories never work.

The Scary Bond Bear Market Story

All we need to come up with a good bond bear market scare story is to examine common analytical assumptions. If we amend these assumptions, we have a story that is possibly as scary as the cinematic classic, The Bloodsucking Monkeys of West Mifflin, Pensylvania.

In the financial markets, it would be safe to say that it would be easy to find commentators that will endorse both of the following scenarios.

1. If the central bank hikes interest rates, it will tend to depress growth, and hence inflation. (The exact transmission mechanism varies based on economic views.)
2. If government debt gets "too large," bond yields rise, and then there is a fiscal meltdown scenario (leading to hyperinflation if the commentator in question likes invoking hyperinflation).

The interesting part of these two views is that they are contradictory: the first implies that rising bond yields suppresses inflation, whereas the second implies that they raise inflation. This is not a bug, it is a feature. Financial market commentators need to jump back and forth between bulls and bears rapidly, and need to have strong opinions regardless of what side of the market they are on. Embracing contradictory concepts means that they have a story to justify whatever their current view is.

In order to generate a more plausible secular bond bear market story, we need to dig into these contradictory views. One of the advantages of Modern Monetary Theory (MMT) is that the theory has dug into these assumptions, as opposed to conventional economics, where the first view (interest rates reduce inflation) is assumed to be true, and there is no questioning of that assumption.

All we need to do is to question the efficacy of rising interest rates to slow economic growth. (It should be noted that Warren Mosler has pushed the following logic the hardest; it may not represent the consensus of all MMT economists. What I am writing here is a paraphrase of Warren Mosler's statements over the years.)

If the policy rate rises, bond yields will also rise. This will imply a greater interest outlay by the government (on a lagged basis), as a considerable part of government debt is relatively short maturity. Furthermore, it raises the interest costs for the business sector, which is a net borrower. Conversely, the household sector is a net saver, and a lot of household borrowing is in the form of mortgages, which are largely fixed in the United States. (Other countries do not have 30-year fixed mortgages, so the interest cost adjusts more rapidly.)

If the household sector has a relatively stable propensity to consume out of interest income, the net result of rising interest rates is to increase household consumption. Rising consumption raises capacity utilisation, and that will likely be more important than the effect of interest rates on investment decisions. The bottom line is that rising interest rates may end up stimulating the economy, for reasons that are similar to the second story above.

However, the key is that this effect is relatively weak. The business cycle is not greatly affected by interest rates (absent the key possibility where a real estate bubble is crushed by higher interest rates).

If policy rates are not particularly potent tool, their precise level is an arbitrary decision of the central bank. This is completely unlike mainstream theory, where the economy spirals to hyper-inflation or hyper-deflation if interest rates are not automatically adjusted to achieve price stability.

In other words, the natural real rate of interest is a chimera. If the central bank thinks the real natural rate of interest is 2%, observed real rates should average 2% across the cycle. If it think the natural rate is 3%, the average will be higher -- with no observable difference in outcomes.

If we accept these premises, generating a self-reinforcing bond bear market is straightforward. A change in personnel at the central bank can result in a change to the central bank's reaction function; it could effectively target a higher natural rate of interest. Rising interest rates raise interest income, raising nominal demand. The resulting higher inflation will cause the more-hawkish central bank to keep hiking interest rates.

The only thing that stops this "doom loop" is the tendency of financial markets to blow themselves up. So long as the central bank does not get too aggressive, there is little reason for rate hikes to derail growth. Demand is rising, and although there are pockets of nuttiness in risk markets, private borrowing has been tepid so far this cycle.

In summary, all we need for a secular bond bear market is for Fed policymakers to stop panicking at the first whiff of a slowdown, and to resume rate hikes more rapidly after a recession. Once the pattern of cutting more in a downturn than during the expansion is broken, interest rates will be able to once again take an upward trend.

Is This Plausible?

Although this scare story is more plausible than others, there are still weak links.

Firstly, private sector balance sheets are heavily encumbered with debt. Unless wage growth is quite strong, debt service concerns will limit how far rates can rise. That said, a secular bond bear market would take place over at least a decade (by definition), and so there will be time to adjust.

Secondly, the tendency for the Bank of Japan to keep rates near zero acts an attractor for developed country interest rates. Hiking rates back to 5% again (for example) would create a huge carry differential, and risk pushing the yen to deeply undervalued status.

Concluding Remarks

I am certainly not calling for a secular bond bear market. That said, a change in the Fed's reaction function as a result of changing personnel poses an obvious risk to be monitored.

(c) Brian Romanchuk 2018