Stay informed with free updates
Simply sign up to the US interest rates myFT Digest — delivered directly to your inbox.
These are treacherous times for bond traders. A couple of months ago, the market consensus was that the US Federal Reserve would cut interest rates six times this year, starting imminently. By Wednesday, however, investors had slashed their expectations so dramatically that many now expect cuts to be delayed to November. Indeed, a slew of higher-than-expected inflation data prompted Lawrence Summers, former Treasury secretary, to warn that the next move might even be up — not down.
Cue hand-wringing from those sectors that have boomed amid ultra-low rates. (A fascinating new report from the IMF report suggests that private capital is one). And cue more speculation about whether expectations will change again this year.
But as this debate rages, it is worth stepping back for a moment to also think about the long sweep of our financial past. And no, I do not mean “history” as traders usually experience it on a trading screen — namely, the late 20th century — but instead, and more thought-provokingly, the past eight centuries. A trio of economists — Kenneth Rogoff, Barbara Rossi and Paul Schmelzing — have been amassing global data on interest rates and inflation since 1311, five decades after Venice started to issue so-called “consols”, arguably the first example of long-term sovereign debt.
Their conclusions were released in preliminary form almost two years ago. But they have now been updated with new historical information, underscoring two fascinating points.
First, you cannot understand the political economy simply by looking at short-term rates, as most previous analyses have tended to do. To be fair, historians hitherto adopted this focus because historical data on short-term rates was more readily available, and 20th-century central bankers wanted to determine the so-called “natural” rate against which to set short-term policy rates.
However, Rogoff et al argue that while the patterns in short-term rates are noisy, if you look at long-term real rates (that is, nominal rates adjusted for inflation) there is a clear and striking trend. These have been steadily sinking over the centuries. They calculate this decline equates to almost 2 basis points a year, on average, since 1311.
The chart is certainly not smooth. Two big inflection points occurred during the 14th-century Black Death pandemic, and then the European “Trinity” financial crisis of 1557. There were smaller inflections in 1914 and 1981.
But what is more striking than these inflections is how rare they are. While long-term rates have often moved in response to recessions, defaults, financial shocks and so on, they almost always revert to trend after a decade or two. As the economist Maurice Obstfeld has pointed out, the result is that they look like mere “blips” from a long-term historical point of view.
To put it another way, modernity triggered an inexorable decline in the long-term price of money, and was doing this well before we started to fret about ultra-low rates in the 21st century.
Why? Previously economists have blamed this on issues such as productivity, demographics and capital flows. Ben Bernanke, former Fed governor, famously pointed to a savings glut in China and elsewhere, while Summers worries about an era of secular stagnation.
However, even more interesting (and counter-intuitive) is Rogoff et al’s failure to find a statistical correlation between real rates and fundamental economic trends. That might reflect the limitations of their data, but the trio offer another explanation. The real reason, they say, for falling borrowing costs is not economic shifts, but an issue economists often ignore — the nature of finance. A combination of modern capital markets, risk analysis and innovation around using collateral to back loans has made money more efficient.
Proving this is hard, but the idea rings true to me. Call this the “against the gods” effect, to cite Peter Bernstein’s seminal book of the same name. A key distinction between modern and premodern societies is that innovations ranging from double-entry book keeping to computers have left us believing that we can predict, manage and price future risks, without relying on gods, as our ancestors did.
In reality, this confidence is all too often misplaced. But justified or not, the cultural shift that accompanied it has made money more abundant and fluid, thus cutting its cost. This is good news. But it also raises two further questions. Will this downward trend ever end? And what does it mean for current rates?
On the first point, the answer rests on your level of imagination. It is hard to believe the trend will continue much further, but it is also difficult to discount future technological advances. Artificial intelligence, say, could increase monetary efficiency.
On the second point, however, the implications are clearer. Adopting an eight-century timeframe suggests that the ultra-low rates we saw in the early 21st century were a slightly excessive deviation from the trend. It should thus be no surprise that long-term rates have corrected upwards, particularly given that the short-term “natural” rate has probably risen.
But this long sweep also indicates that what is happening now is not remotely unusual. Just don’t tell that to the bond traders who are feeling bruised by recent events.
gillian.tett@ft.com