US debt is risky: A rant about evidence-based investing

A couple of months ago, Hanno Lustig presented his latest research paper at the Jackson Hole conference of central bankers. The paper got quite some media traction and was written about as a revelation by the Economist, Reuters, and others. In my view, this simply shows how ill-informed many people are and how they are simply parroting conventional wisdom without checking it empirically. But one thing after another…

First, let me state that the paper by Hanno Lustig and his collaborators is excellent. But that is no surprise. Almost everything that Hanno does is excellent research. In this instance, I have written about a previous version of the paper he presented in Jackson Hole in my post on 3 July.

The researchers looked at US Treasury yields during the Covid crisis. To cut a long story short, he found that on days when the US government introduced spending decisions to rescue the economy, real yields increased, driving bond prices down. The chart below summarises the key findings of the paper.

Whenever the US government announced deficit spending measures, bonds declined, and the drop accelerated during the pandemic when unfunded deficits rose rapidly. Meanwhile, on Fed meeting days, bond prices followed the rate decision and the guidance of the Fed with yields dropping when the Fed cut rates or introduced QE and yields rising when the Fed increased policy rates.

Cumulative change in valuation of US Treasuries

Source: Gomez-Cram et al. (2024)

So, this is direct proof that US Treasuries are not risk-free. Instead, whenever the US government increases its budget deficit without paying for it with other means, the market increases the risk premium on US Treasuries to account for the possibility of default.

But why is this news? To me, it is obvious that US Treasuries are risky, and I think you must be entrenched in some version of the myth of American exceptionalism to believe that US Treasuries are risk-free.

Recently, the CFA Institute surveyed 4,243 members about the future of the US Dollar. Apart from believing in the usual trope that the US Dollar is at risk of being replaced as the world’s reserve currency (something that I have addressed as being far from the truth here, here, and here) they apparently aren’t worried about US Treasuries either.

77% of respondents say that US public finances are not sustainable and only a minority of 20% said they are (Personally, I think they are not sustainable). When asked if the US government will be able to reduce its debt/GDP-ratio over the next 5 to 15 years, three out of five respondents (61%) said no. So, the vast majority of respondents think US debt is not sustainable and going to rise over the next 5 to 15 years. Yet, when asked if investors are losing confidence in the ability of the us to borrow to fund its obligation, only 34% said they do. Again, roughly three out of five respondents (59%) said, they are not worried about investors losing confidence in Treasuries. So CFA charterholders on average think that the US is living above its means and has too much and growing debt but expect investors to continue to hand over their savings to fund the deficit, no questions asked. I wonder what the they would say if we replaced the US with Italy or Brazil?

I think what these media reports and surveys show is that there is a large group of people in the world who simply take certain conventional pearls of wisdom for granted. Among them are tenets like ‘US Treasuries are risk-free’ or my all-time favourite bugbear that ‘printing money creates inflation’. Other pieces of conventional wisdom that are simply not true are ‘stocks are a hedge against inflation’, ‘in the long run stocks are less risky than in the short run’, ‘stocks outperform bonds in the long run’, or ‘higher taxes are bad for the stock market’.

Once you try to empirically verify each of these conventional wisdoms you very quickly end up frustrated because they simply aren’t true in real life even though they sound true in theory.

This is why I write these daily posts. As I say in the very first post every year, I am an empiricist, and I don’t believe in any economic or finance theory unless you show me evidence that it works and how it works. I have simply lost too much money in my career because I invested based on some theory that said markets should do this not that.

But unfortunately, evidence-based investing is still a niche endeavour. All you need to do is read newspapers or watch pundits on TV talk about financial markets and the economy. The vast majority of them don’t provide research insights, they provide commentary.

They never substantiate their views with anything other than a few charts, most of which don’t even give you a basic quantitative assessment of how important a specific development is. That is probably why people still think that earnings are more important for share prices than bond yields when in truth it is the other way round (see here and here).

If you are a consumer of investment research or economic research, especially if you are a professional, you always need to ask analysts to substantiate their views with a quantitative analysis. This analysis needs to show that there is at least a correlation, or better yet causation (but that is often impossible in practice), and how important the development is economically. I can give you a ton of risk factors and scenarios that may well happen, but they are either so unlikely that it is a fool’s errand to invest based on them, or if they happen they have such small effects that you will never be able to make money off of it.

This brings me to the final point in this admittedly long and increasingly angry rant. Some people have argued that the new research by Hanno Lustig shows that the US has too much debt and that unless the deficit is reduced, this debt bomb will lead to a bond market crisis and the eventual default of the US.

Utter nonsense if you ask me.

If you read the paper, you see that the cumulative increase in real yields in US Treasuries between March 2020 and November 2023 was a whopping 0.53%. And that is the combined effect of a higher default risk premium and higher real rates reflecting higher economic growth in 2023 compared to 2020 as well as a ton of other factors. The largest unfunded deficit increases in the history of the US to the tune of $5.43 trillion(!) dollars (26% of US GDP) and all we have to pay for it is an additional 0.5% in interest or less? Are you kidding me? That is the bargain of the century.

Another thing you will find out when reading Hanno Lustig’s paper is that the impact on risk premia for Germany and the UK was larger than for the US. In other words, default risks may rise if the US deficit increases, but thanks to the dollar’s status as the world’s reserve currency, the US can increase both debt and deficits much more than the UK or Germany can before it experiences the same increase in debt costs.

And guess what, we have the example of another country that does not benefit from global reserve currency status and has managed much larger deficits and debt than the US has for thirty years without any problems. That country is called Japan.

Finally (and you better imagine me with a bright red head and steam coming out of my ears because that is how I look right now…), I haven’t even begun to mention the Fed and its ability to simply sweep up all the excess government debt and hold it on its balance sheet forever to prevent bond yields from rising too much to do major harm. How this can work, is explained here and here.

In short, if you listen to pundits who do not substantiate their views properly, you are bound to lose money and make major investment mistakes. As I said above, I have lost too much money in my life listening to these people (I call it school fees paid for the University of Life) but at some point, we all should wise up and change our habits and simply ask whoever is giving us advice about this or that a simple question: “Can you show me why you think this is going to happen and how big the effect is?” The answer you get will tell you if you should listen to these people ever again.

And now I need to cool off and take a break before I start mentioning the names of some useless pundits…