AQR doesn’t publish short-term forecasts for asset prices, so you won’t find a 2020 outlook page here. We do, however, look at others’ predictions, often to point out how inaccurate we believe they were. It’s easy to find fault in others’ forecasts because events seem far more obvious in retrospect. 1 1 Close Who could have predicted that Ricky Gervais would insult the audience at the Golden Globes? Sometimes we feel like that annoying kid who laughs at all the wrong answers in class without raising his hand to answer a question himself. However, as difficult as forecasting is and as poor as people’s ability to predict the future may be, forecasts can still be useful for investors. We use forecasts in some of our models and are aware of the importance investors place on longer-term forecasts in asset allocation decisions.
We’ve done many quantitative studies of forecasts, but we thought it might be helpful to take a slightly less scientific look at predictions from last year to see how they played out. We compiled a broad set of forecasts from economists at large institutions. In addition to their guesses on economic data and markets, we looked at their more qualitative outlooks, i.e., the ones with words. Using a sophisticated natural language processing algorithm (one of our researchers read them) we broke them down by theme and direction. Some of the predictions seemed like they came from Monday Night Football commentator Booger McFarland, such as “As long as China’s all-important demand for commodities holds up, cyclical commodities should remain supported.” 2 2 Close Booger is known as the master of the obvious. For example, he provided the analysis, “It’s a run/pass option meaning they have the option to run or pass it.” Good call. We decided to make a play and not include this “view” or any others which were devoid of content.
Many of the outlook pieces did contain actual market views. We found that they tended to be bearish on bonds and credit but were bullish on commodities. There was a wide range of views on stocks, which is slightly unusual in that strategists tend to be bullish stocks, for the obvious reason that stocks tend to go up. 3 3 Close Like Booger’s comment “Gotta go for two right now, because going for two would make it 19-16, a 3 point game." There were several common views about the economy. The first was that both economic and corporate earnings growth would slow. Few forecasted a recession, but “late cycle” was a common phrase. On the inflation front, there was a general belief that prices would rise a little more quickly than they did in 2018, and that oil prices would rise modestly. The mix of weaker earnings and higher inflation explains the negative calls on credit and bonds. Many analysts recommended defensive high-quality investments.
If we think back to the end of 2018, the predictions make more sense. The stock market was coming off of its worst year since the crisis. The Fed had hiked four times during the year, which had led to a sell-off in bonds. Many investors and economists feared the Fed had made a policy mistake. In that environment, it would be challenging to make bullish equity forecasts, though a few analysts did. It was much safer for a strategist to recommend being defensive.
As we all know now, equities and credit had exceptional returns in 2019. 4 4 Close Per Bloomberg, 2019 total returns for the S&P 500 index and the Bloomberg Barclays Global Agg Credit Index were 31.5% and 10.7%, respectively. Bonds were also up. 5 5 Close The Bloomberg U.S. Treasury Index delivered a total return of 6.9%. The forecasters (in aggregate) did not get any of this right. But some of their less prominent calls were better. We don’t have the final numbers for 2019, but in all likelihood economic growth was slower than it was in 2018. 6 6 Close With most major 2019 data in the books, economists polled by Bloomberg see 2019 growth at 2.3% vs. 2.9% in 2018. Energy prices rose as did some broader measures of commodity prices such as the Bloomberg Commodity Index. Inflation also rose as many had predicted, though it was still low in an absolute sense. 7 7 Close December CPI data will be published next week, and economists polled by Bloomberg expect a YoY increase of 2.3%, compared to 1.9% in the year earlier period. To sum up the results, the forecasters got the economy right but the markets wrong.
Now after a good year of returns for risky assets, forecasters are a bit more optimistic. Many are predicting another year of good equity returns. Of course, there are still bears, but their numbers are thinner. Analysts are also cautiously positive on bonds and credit. They generally prefer risky assets to safe ones. This probably has something to do with the positive returns from last year. Few expect high inflation, and many are more optimistic on growth.
Where does this leave us? It seems that forecasts tell us more about the past than the future. They are products of recent events in the economy and markets. It might seem that investors therefore should ignore forecasts or perhaps do the opposite of what the forecasters tell them. While we are sympathetic to this view, it may be a bit too extreme. Even if predictions are anchored to the past and slow to change, they can still be useful because the markets exhibit many of the same characteristics. While the forecasts last year may have been flawed, markets also got some things wrong, such as the Fed and inflation. 8 8 Close You don’t always have to be right, just more right than markets. It’s the famous “I don’t have to run faster than the lion, I just have to run faster than you.”
Forecasts also exhibit certain behavioral patterns. Like markets, they can be serially correlated – when they start moving in one direction they tend to continue in that direction. It is more important to follow how the forecasts evolve than to expect them to be accurate at any point in time. Last year, for example, forecasts for inflation fell over the course of the year. So even though the initial calls were inaccurate, these changes indicated better returns for stocks and bonds later in the year. So forecasts can be used to evaluate market pricing and as a momentum indicator. But let’s not take this too far. Forecasts can be useful in models but should not lead to oversized, concentrated positions. You probably could have forecasted that I would say that.
What We Are Watching
U.S. CPI (Tuesday)
U.S. CPI was unexpectedly subdued through most of 2019, defying expectations that a tight labor market would lead to higher inflation. Over the last few months, CPI readings have firmed somewhat, with both headline and core inflation moving above 2%. 9 9 Close Bureau of Labor Statistics. Historically, the current combination of low unemployment and rising inflation would likely have led the Fed to consider tightening policy. However, following several years of below-target inflation, policymakers at the Fed have signaled they are unlikely to hike rates unless price pressures increase significantly over a prolonged period. 10 10 Close In his most recent post-meeting press conference, Chair Powell stated that “we would need to see a really significant move up in inflation that’s persistent before we would consider raising rates to address inflation concerns.” This stance may reduce market sensitivity to upside surprises in inflation data in the near-term. In contrast, lower-than-forecast readings might open the door to additional rate cuts if downside risks to growth become more prominent.
U.S. Retail Sales (Thursday)
Consumer spending growth slowed somewhat in 2019 but remained the largest contributor to U.S. GDP growth. This week, the Census Bureau will publish retail sales data for December, the most active month of the year for consumer spending. Retail sales grew at a sluggish pace in November, and a soft December reading would indicate cautious behavior by households in the holiday period. A pullback by consumers would be a bearish sign for the U.S. economy, especially in light of recent weakness in capex activity. Such a result might boost U.S. fixed income, while weighing on the dollar and equity prices.
China GDP, Industrial Production, Retail Sales (Friday)
The Chinese economy decelerated notably in 2019 in response to rising trade tensions with the U.S. and prior policy actions aimed at reining in excess leverage in the country. Chinese GDP growth fell to multi-year lows in the third quarter and economists surveyed by Bloomberg expect it to have remained near those levels as of year-end. With that noted, higher frequency data such as industrial production and retail sales have recently been showing signs of stabilization, surprising consensus forecasts to the upside for November. Optimism around trade negotiations and Chinese stimulus may be starting to show in the data. If maintained in December, these improvements may presage a brighter outlook for the start of 2020. Positive surprises in Chinese data next week would support risk appetite, likely driving equities higher and weighing on bonds.