We face a myriad of economic data and conflicting expert opinions everyday. A part of me has always craved ever a modicum of understanding of that subset of human drama. Despite the less than stellar track record of institutional economists, I have always thought trying to stay ahead of economic cycles a noble endeavor.
So the choice of this book for my first review was an easy one. As the full title suggests, “Ahead of the curve: A commonsense guide to forecasting business and market cycles” is about the reliable leading indicators to business and market cycles. Joseph H. Ellis was a partner at Goldman Sachs and the top ranked retail industry analyst for 18 years. His real-world, result-oriented experience shows where he let actual data and their inter-relationships instead of formulae do the talking. The result is a very convincing and readable treatise that an amateur like me could follow.
Let me try to summarize the key approaches and findings of this book.
* Economic data are noisy and subject to seasonality effects. The best way to visualize them is to plot them on a year over year (YoY) change basis. The fluctuations in monthly data are smoothed out with three month averages. Ellis called this focus on relative change rather than absolute levels the ROCET (rate of change in economic tracking) approach.
* Ellis uses a very practical definition of economic causality: 1) Plausible connection between the cause and effect; 2) Clear indication of the cause leading the effect when viewed through the ROCET approach.
* Consumer spending (Personal Consumption Expenditure or PCE) is THE leading indicator of business and market cycles. Bear market begins when real PCE slows down.
* Industrial production, real capital spending and employment are lagging indicators.
Casting aside the lagging indicators is every bit as crucial as pointing to consumer spending (real PCE) as the leading indicator. It is hard not to pay attention to real capital spending (commercial real estate is hot!) and employment (the unemployment rate is 4.7%!), but Ellis clearly showed that it was more the rule than the exception that they peak well into a bear market!
Equally illuminating is his treatment on “asymmetrical circular causality” a term he applied to the causation circle between real consumer spending and employment:
Jobs (employment) and the income created by them are key elements in consumer spending power. Yet consumer spending, representing approximately two-thirds of GDP, in turn drives the demand cycle for industrial production, services, and capital spending, which are cornerstones of employment. The question is, which leads which? The answer is that consumer spending, cyclically speaking, is more powerful in driving employment than vice versa.
The same “circular yet lopsided” causality exists elsewhere, most relevantly in commodity prices and economic activity. It is my contention that the level of economic activity is a far more rigorous driver of commodity prices than the inverse because of other inputs involved; however, the circular relationship means any perma bull (or bear for that matter) can interpret any development to suite his point of view. A very good case in point is illustrated by Barry Ritholtz of the Big Picture.
A special website, www.aheadofthecurve-thebook.com, is devoted to this book where can find color versions of the charts in the book, updated through the beginning of this year.
Chart 11-3 shows the lagging relationship between employment and consumer spending:
Because the charts at www.aheadofthecurve-thebook.com are not totally up to date, I plotted PCE using the same method below. The decline started in April has continued into July. I have been writing about a housing-led slow down , so this does not surprise me.
There are certainly caveats here, one of which is that a decline in YoY PCE is not a sufficient condition for a bear market as a close reading of the charts will reveal. For now, the market seems to be brushing aside the possibility a slow down. Personally, I'm not so sanguine. Time will tell if PCE remains a good predictive tool in this cycle.