As quantitative investors, we like to view the world through numbers. We wish we could assign probabilities to any possible future event, but markets don’t always lend themselves well to such analysis. A price can represent an almost infinite combination of variables, so we have to avoid drawing easy conclusions. One exception would be short interest rates, which seem to price the very discrete actions of central banks with amazing precision. The financial news is full of stories explaining how the futures are pricing a 36.423% chance of a specific Fed move at an upcoming meeting. Sometimes they add additional complexity in the name of clarity by creating probabilities of multiple outcomes. They are able to tell you that there is a 47% chance of a 25bp cut, and a 7% chance of 50bp cut, but also a 3% chance of a 25bp hike. 1 1 Close Sometimes the numbers don’t add up to 100%. There is no explanation as to why. These articles might be useful if you have a highly-levered book of short-rate instruments or if you really enjoy guessing bond yields, but most investors have more general questions about monetary policy. Questions like: are rates likely to go up or down in the next few years?
These questions shouldn’t require complex analysis. Fed policy tends to be persistent and, in theory, well communicated. 2 2 Close At least they try. Sometimes. It would be unusual to have a chaotic rate path: a cut, followed by a hike, followed by another cut. While central banks sometimes surprise investors, these shocks are limited and calculated exceptions. That is why many folks are puzzled by current market pricing. The Fed has been hiking for several years, so the law of inertia says hikes should continue. Some economists are forecasting the Fed to be on hold for the rest of the year. Fed statements have signaled a move to neutrality, but have not given a strong indication that rate cuts are coming soon. 3 3 Close The Fed’s March 2019 Summary of Economic Projections showed the median Fed forecast for the federal funds rate increasing from 2.4% in 2019 to 2.6% in 2020 and to 2.8% in the longer run. Source: Federal Reserve. It would seem unlikely in such a scenario that short-term interest rate markets would price rate cuts, but they are. 4 4 Close We won’t even try to tell you the distribution of the odds. This split between economists and prices is an anomaly, but we think it is consistent with some trends we’ve seen in recent years.
The concept of term premium is helpful for figuring out why things are where they are. Many economists have noted that long-term bonds often yield more than what people think rates will be over the same period.
One of these economists might even work here. And be named Antti Ilmanen.
For example, if investors think the short rate will average 3% over the next 10 years, they may demand 3.25% for a 10-year bond. Economists call this extra yield “term premium” and have several theories as to why it exists. One of the most cited reasons is that it reflects the cost of liquidity: investors require compensation for giving up the use of money for such a long time.
In the case of U.S. government bonds, credit does not have much impact on pricing.
This shouldn’t apply as much to short rates, where there is less of a term to have premium, so we have to consider different explanations for its existence in shorter maturities.
There are multiple ways to measure the differences between short rates and forecasts. We have our own proprietary models, but for the purpose of this piece, we’ll use the Fed’s two-year term premium measure. The Fed researchers employ a somewhat complicated five factor model, but we’ve found that the results are not that much different from simpler models. 7 7 Close For a detailed look at the construction of the Fed’s term premia model, see: Tobias Adrian, Richard K. Crump, and Emanuel Moench, "Pricing the Term Structure with Linear Regressions," Journal of Financial Economics 110, no. 1 (October 2013): 110-138.
As you can see for most of the 70s, 80s, 90s and whatever the decade after that is called, 8 8 Close 00s? Aughts? Millennial?? the term premium was positive. This means that the rates priced by markets were higher than the rates implied by forecasts. For example, if forecasters said there would be two hikes in a given year, the market might price three hikes. Then, around 2012, it reverses. For the first time since the early 60s, we’ve seen a sustained period of negative term premium. For the past few years, the markets have persistently priced lower rates than what economists have forecast.
The shift in term premium didn’t happen suddenly: it has been gradual and seems to loosely coincide with falling rates. The shift is probably a combination of changes in the bias of both pricing and forecasts. There are several potential reasons for forecasters to bias higher in recent years. As rates have fallen, they may have kept to their old models or the belief that rates will go back to previous levels. They may have been thrown off by unconventional monetary policy. Maybe forecasters have just been wrong. 9 9 Close Probably not. The evidence on their accuracy is mixed. It’s difficult to say conclusively that the forecasters are getting worse. While they have been high at times, they were more accurate than market pricing last year.
There are some economic reasons that the bias in pricing may have changed. For many years following the 70s, businesses perceived inflation and higher interest rates as two of their biggest risks. They had experienced its effects firsthand. They may have used the short-term interest rate market to offset some of that risk. The market was providing a premium to give this protection through higher yields in short-term bonds. As inflation and interest rates started to come down, inflation may have seemed less menacing. Then the financial crisis and the accompanying recession were so severe that economic weakness started to seem like more of a menace than inflation. Perhaps current short-rate pricing reflects the desire for insurance against recession rather than inflation.
This shift is important for investors beyond short-rate traders. 10 10 Close We do use this in some of our alpha signals. You know we are all about risk premia. Many people look to the front end of the yield curve for signals on the health of the economy. If rate cuts are priced in, investors may think that indicates a future recession or that a very reactive Fed will come and save the stock market. These signals should be viewed with caution. The market may be telling us more about investor preferences than the economy. The shift is also instructive for understanding risk premia in general. While many folks think risk premia are permanent, it is possible for them to change over time. Because there is noise in the short run, it can be difficult to see. However, if you understand the reasons for the risk premium to exist and what might change it, then some of the most puzzling market pricing can start to make sense.
What We Are Watching
E.U. Parliamentary Elections (Thursday May 23rd through Sunday May 26th)
From Thursday to Sunday, the 28 member states of the E.U. will hold elections for the European Parliament. The authority of the European Parliament is somewhat limited, as major decisions require coordination with the Council of the E.U., which directly represents member state governments. However, the elections provide an important update on political trends at the regional and national levels. Across the E.U., recent years have featured an erosion of support for traditional centrist parties in favor of unconventional and often euro-skeptic newcomers. Continued movement in that direction could be negative for market sentiment, particularly if the two largest centrist coalitions (the EPP and S&D) underperform, complicating the formation of a centrist majority bloc.
At the country level, markets may be most sensitive to electoral results in Italy, which has been governed for nearly a year by a coalition of two anti-establishment parties: the Five Star Movement (M5S) and Lega Nord (Lega). While M5S is nominally the senior member of the coalition, opinion polls have shifted strongly in favor of Lega over the last year. If E.U. parliamentary voting delivers a strong enough result for Lega relative to M5S, it could trigger the breakdown of the coalition and a new general election in Italy.
Bank of Canada (Wednesday)
At its last meeting, the Bank of Canada (BoC) held its policy rate unchanged at 1.75%, but post-meeting communications were more dovish than expected, downgrading its 2019 GDP growth forecasts and removing wording that previously suggested a tightening bias. However, in his post-meeting press conference, Governor Poloz stated that interest rates “are more likely to go up than down.” 11 11 Close Bloomberg: “Bank of Canada Abandons Rate-Hike Bias Amid Economic Slowdown,” 4/24/19. A few weeks later, Poloz reiterated this message, saying: “the natural tendency is for interest rates to still go up a bit.” 12 12 Close Bloomberg: “Poloz Sees Higher Rates as the ‘Natural Tendency’ for Canada’s Economy,” 5/17/19. Although no action is expected from the BoC at this week’s meeting, the market will look for clarity on the BoC’s expectations for policy moving forward. While Poloz continues to suggest the next move is likely to be a hike, market pricing would suggest the next move is likely to be lower.
China Manufacturing PMI (Friday)
China’s CFLP Manufacturing PMI declined in April after staging an encouraging recovery in March that took the index back into expansionary territory. The April decline suggested that seasonal factors, specifically normalization in activity after the Lunar New Year holiday, may have boosted the March data. This was a disappointment to investors expecting improvement in growth following fiscal stimulus measures implemented by the Chinese government. With the PMI only marginally in expansionary territory and the trade war keeping investors on edge, the May release will be closely watched to assess the health of the Chinese economy.