As the S&P 500 Index teetered on the verge of a bear market at year-end, we fielded numerous questions about the prospects for an impending recession. Our answer is that we do not expect a recession in 2019. We will devote this quarter's update to explain why.
We are students of economic and financial market history, and we believe only a small handful of indicators help to forecast a recession. They are the slope of the yield curve, monetary policy settings, the supply of credit to the real economy, and clear signs of misallocation of capital. We would argue that none of these are present today.
The yield curve
The slope of the yield curve plays an important role in the economy, and we can demonstrate this both theoretically and empirically. The Swedish economist Knut Wicksell first theorized in the early 20th century the idea of a natural or neutral rate of interest that would result in price stability. In other words, it would be neither restrictive nor stimulative to the economy. In subsequent decades, the Austrian school of economic thought developed this idea further, and while the neutral rate remains theoretical and unobservable, to this day it plays an important role in formulating monetary policy. Monetary authorities (such as the U.S. Federal Reserve or Bank of Japan) directly influence the front end of the yield curve relative to the neutral rate. The front end of the yield curve is also a reasonable proxy for the cost of capital in the economy because it directly influences the cost of money in the banking system. This includes the rate at which the monetary authority lends to banks, the rate at which banks lend to each other, and the rate at which banks lend to firms.
On the other hand, the long end of the yield curve is a rate set by the market, and as such, is a reasonable proxy for the return on capital. Yields on the long end of the yield curve represent the rate at which the bond market is willing to lend to firms for long-term assets such as capital equipment or factories (plus some spread compensation for default risk). Presumably, firms would not be willing to borrow at such a rate unless they considered the investment to be profitable net of that borrowing cost. So the slope of the yield curve is essentially a spread between a proxy for the return on capital and a proxy for the cost of capital. This explains the basic theory of why the slope of the yield curve matters.
The theory also holds up to empirical scrutiny. The best and most recent summaries of the forecasting power of the yield curve come from the San Francisco Fed, which has published two research notes: Economic Forecasts with the Yield Curve and Information in the Yield Curve about Future Recessions. A positive spread between the long end and short end of the yield curve (which we interpret as maturities of up to two years) — even if only slightly positive — is indicative of a reasonably well-functioning economy. That is where we find ourselves today.
The policy rate
The second indicator that we monitor is the real policy rate, which indicates whether the monetary authority is loose or restrictive. In the United States, the corresponding metric is the real federal funds rate. It is at this rate, relative to inflation, that banks lend excess reserves to each other. The Fed's preferred inflation measure is the year-on-year change in core personal consumption expenditures price index. Note that the real federal funds rate had been negative (that is, set below the PCE inflation rate) for most of the post-2008 global financial crisis period. In that period, a negative level appropriately signaled the necessary accommodative monetary policy as the economy was recovering and the unemployment rate was declining. By comparison, in the preceding handful of market cycles, the real federal funds rate had to exceed 200 basis points to exert any kind of slowing effect on the economy.
In the current cycle, the real federal funds rate ticked slightly above zero only in the second half of 2018. In the final days of 2018, the rate was only 60 bps — far below the restrictive levels of previous cycles.
Availability of credit, the lifeblood of the corporate sector, is another instrumental signpost leading into recessions. To track this indicator, we closely monitor the Fed's quarterly Senior Loan Officer Survey for hints that the banking system might be tightening the availability of credit. While this survey doesn't have as much history as we would generally prefer (it began in 1990), its theoretical underpinnings are solid, and it has given an accurate warning ahead of the past three recessions. It showed signs of stress in 2016. A technical recession did not occur, but we would describe this period as a mini-recession. The stress in the energy sector resulting from the collapse in oil prices from the third quarter of 2014 to the first quarter of 2016 caused an uptick in the default rate in several important sectors of the economy.
Today, the loan officer survey shows that we begin 2019 with ample evidence that banks are willing and able to lend.
Finally, signs of malinvestment or gross misallocation of capital are admittedly harder to define and thus defy tracking by a simple quantitative measure. Here, we instead must draw upon our combined experience and take some comfort in the fact that "we've seen this movie before." One example is from the days of the tech bubble in 1999. It was fairly clear to us then that blank-check initial public offerings (IPOs) in the tech sector, which gave companies high valuations despite hardly any revenue, and traded on ratios defined by "clicks" or "eyeballs," represented a misallocation of capital. Similarly, when median home prices relative to median income levels exploded upward in 2005 and 2006, we felt confident that it was unsustainable. In contrast, today we struggle to find such clear cases of large pools of poor capital investment decisions that could cause a recession.
Trump, trade, and the re-election goal
We end by observing that the risk to the economy now would clearly come from a retreat of animal spirits in general. This could result from a prolonged trade war with China. So far, the NFIB Small Business Confidence Index and the various Conference Board measures of Consumer Confidence continue to suggest that both firms and consumers are in good health. We can only assume that President Trump wants a second term, and it should be clear to the administration that a prolonged trade war would be the proximate cause of recession should one emerge at this stage. A trade war would therefore represent an unforced error at a critical time when the 2020 race would just be getting started.
In sum, we view the current environment as being most similar to the sell-off in risk assets in mid-2011 during the European sovereign credit crisis. The global economy at that time was in the midst of a mid-cycle slowdown combined with policy uncertainty regarding the stability of the euro. We held our ground on portfolio positioning then, and while it took a couple of quarters for the market to make new highs, eventually pricing and fundamentals re-converged. We believe investors will be best served today by standing firm and not responding to fear.
More from Capital Markets Outlook
See our asset allocation preferences