“Peter Cook is the author of the ‘Is That True?’ series of articles, which help explain the many statements and theories circulating in the mainstream financial media often presented as “truths.” The motives and psychology of market participants, which drives the difference between truth and partial-truth, are explored.”
On a day like today, when the holiest of all data points is released (employment situation), it may seem strange to focus on a wider picture. But we must remember that no single data point can describe the economy at any point in time, let alone describe its future trajectory. For example, even if the payroll data showed a monthly gain 1,000 or 500,000, the divergence from the recent tendency of 200,000 would be so great that it would be written off as an outlier. At least until it occurred again next month. Quite simply, one data point does not a trend make.
If it is difficult to describe the economy, it is even more difficult to describe the consensus view of the economy. The consensus view is presumably baked into market prices at any given moment. Successful investing is based on understanding the difference between reality and the perception of reality, which explains why it is difficult to be a successful investor. That said, it may be possible to detect points at which the gap between the economy and the perception of the economy is at its greatest, which is the subject of this article.
Summarizing Part 1 of this series, a spike in economic activity in late 2017 was largely related to rebuilding in the wake of the destructive hurricanes that hit Florida, Texas, and Puerto Rico last September. The bond market (and the Fed) appear to be extrapolating the GDP spike into the future, pushing up yields in anticipation of higher growth and inflation. For example, TLT, the ETF which tracks long-term bond prices, has declined 6% since early September 2017. In contrast, stocks in the auto, homebuilding, and transportation sectors rocketed higher during Q4 2017. Over the last couple of months most of those gains have been unwound.
The recent decline in the stock market appears to be more closely aligned with reality, but there is another way to understand the debate on reality and the perception of reality. The Citigroup Economic Surprise Index (CESI, shown below) tracks the difference between actual economic data (reality) and economists’ expectations for economic data (perception of reality). If economists are “surprised” by data that exceeds their expectations, the CESI rises. If economists are “surprised” by data that is worse than their expectations, the CESI declines.
CESI oscillates over time as economists alternate between being too optimistic and pessimistic relative to underlying reality. As a result, the most important CESI chart points occur just after an extreme is reached and is subsequently followed by a reversal, because that’s when expectations are too high/low but are beginning to reverse. For example, between CESI’s low on June 15, 2017 and its peak on December 26, 2017, economists were consistently too pessimistic about the economy’s performance. Anyone who had knowledge of that change in mid-2017 could have sold bonds and bought stocks, which would have been profitable. In fact, stocks rose 10.2% between June 15 and December 26, while bonds were flat during that period.
Since CESI’s peak in December 2017, economists have been too optimistic about the economy’s performance. Anyone with advance knowledge of that turning point could have bought bonds and sold stocks. Stocks have indeed declined by 3%, but bonds have declined by even more, or roughly 4%. But this anomaly may just be a matter of timing. That is, the current consensus narrative, “synchronized global growth,” may take more time and data before being discarded. For example, when CESI previously reached rarified air (2011 and 2012), it subsequently reversed all the way into negative territory. There may also be a seasonal pattern to CESI, because it tends to peak during Q1 and then decline until the middle of the year, with an exception in 2016. In early 2018, are we simply witnessing a seasonal bout of excessive optimism?
Other data confirms that the synchronous global growth story is getting long in the tooth. ECRI, which tracks global business cycles, shows that the peak in growth occurred during Q2 or Q3 of 2017. It is always possible for growth to re-accelerate, but it isn’t clear what set of events, and on what scale, would cause a re-acceleration.
Looking into the future, using ECRIs methodology of leading indicators, the chart below shows that economic growth may decline from current levels. The top three indicators are leading indicators of economic activity. When they peaked in early 2014, the bottom blue line (a proxy for current US GDP) began to fall throughout 2015. The leading indicators bottomed during 2015, foretelling a rise in GDP that began in 2016. Most recently, the leading indicators peaked in late 2016 and early 2017, which presaged the peak in GDP in late 2017, and which should be followed by further weakness during 2018.
Many would argue with the idea that economic growth will decline during 2018. That group includes the Fed, which predicts GDP growth of 2.7%, which is well above its 1.9% estimate of long-run US GDP growth. This argument is based upon a belief that tax cuts and heightened fiscal spending will rev up the economy in 2018. After all, leading indicators peaked in early 2017, but they couldn’t have predicted that a stimulative tax cut would occur in late 2017, which is what economists call an exogenous shock. This argument has some merit.
But it is also possible that tax cuts are the reason that CESI is registering excessive optimism, which is now reversing. In this view, a growth impulse from tax cuts will likely be overwhelmed by declining business cycle dynamics plus the well-known headwinds from aging demographics and excessive debt.
CESI is not perfect, but it is a good attempt at measuring the difference between economic reality and the perception of economic reality. CESI reached a peak in late December 2017, possibly because economists extrapolated an unsustainable spike in GDP growth due to rebuilding after the hurricanes. In addition, tax cuts and other fiscal stimulus are perceived to kick-start economic growth into a higher gear. As a result, bond yields shot higher during Q1 2018. But the auto, housing, and transportation stocks, which benefited most from rebuilding activity, have reversed almost all their gains since September 2017. The bond and stock markets appear to be in disagreement.
CESI is most useful after it reaches an extreme and then reverses, because it indicates that excessive optimism or pessimism is in the process of reversing. After peaking in late December 2017, CESI did what it normally does when it reaches such a high level; it reversed. But CESI is still in positive territory, suggesting it could fall much further, as it did after reaching similar heights in 2011 and 2012. CESI also exhibits seasonality, because it tends to peak early in the year, followed by a significant decline into the middle of the year. So the risk that CESI continues to decline during 2018 is high.
Leaving the world of expectations (perceptions of reality), data shows that global growth peaked 6-9 months ago, and leading indicators of future economic growth are pointing downward. These facts undermine the synchronized global growth narrative.
The destruction of the synchronized global growth narrative is probably the main investment risk of 2018. If economic data is falling during a period of high expectations for economic growth, disappointment will surely follow, and CESI will decline accordingly. That is, the “Growth and Inflation Scare of 1H 2018” may be nearing its end.