Contact Us



Interview: Short Recession Call Revisited

The Economic Cycle Research Institute (ECRI) specializes in “calling cycle turning points in economic growth and inflation” and has a long, exemplary history of getting those forecasts right. Cofounded by the late Geoffrey H. Moore, winner of the American Economics Association Distinguished Fellow Award, Lakshman Achuthan, and Anirvan Banerji, ECRI analyzes not the gyrations of the markets but larger business cycles.

In the days following the stock market’s March 23, 2020, bottom, as the world was pounded by COVID-19 and the economy by shutdowns designed to contain the virus, ECRI issued a forecast for “A Nasty, Short and Bitter Recession.”

It is hard to overstate how important and how difficult these forecasts are. We don’t yet know when a vaccine or effective treatments will be made widely available; we don’t yet know how extensive or destructive the next surge of the virus will be; we don’t yet know the full impacts of political decision making or of human behavioral responses such as wearing masks and social distancing or refusing to do so. Nor do we yet know the extent of political and social unrest and their implications. With those variables in mind, here is where Lakshman Achuthan and Anirvan Banerji now see things, two months after their original report.

[David Kotok]


Kotok: We have metaphors in history: the Spanish Flu of 1918, World War 2 deficit financing, Great Financial Crisis (2008–09) monetary expansion, and, tragically, an updated version of the 1968 inner city turmoil, rioting and destruction. Others may be added to the list. My first question: How do you use these strategic and historical metaphors to help you forecast in these extraordinary times? What metaphors are the strongest for you?

ECRI: ECRI’s focus remains economic cycles, and on business cycles going back more than a century and in a variety of market-oriented economies. A key insight is that – across space and time, so to speak – cycles have always been driven by a common set of drivers, regardless of the metaphors, narratives, and shocks surrounding them.

Our indicators also go way back. So we can see how ECRI’s objective leading and coincident indexes behaved around the Spanish flu, World Wars I and II, the global unrest in 1968, and the Global Financial Crisis, as well as the panic of 1907 – the first 20th century recession generally classified as a depression – that ultimately led to the creation of the Federal Reserve in 1913. And we regularly monitor 22 economies including the G7 and the BRICS, so we understand how business cycles have evolved in a wide variety of structural circumstances.

Most importantly, our quest has long been to understand the conditions under which our cyclical framework would fail. We find that good leading indexes – created on a conceptual basis rather than being optimized and back-fitted – continue to work in terms of cyclical direction in a very wide range of circumstances. So, in constructing our indexes, our emphasis has always been on robustness, rather than optimization.

The seven-month 1918–19 recession – which overlapped with the Spanish flu – was relatively mild, but still deeper than the milder post-WWII recessions. However, just nine months after that contraction ended in March 1919, the 1½-year 1920–21 depression took hold. With the Fed raising rates and the government focused on balancing its budget, this became a far deeper downturn – substantially deeper than the 1937–38 recession and at least double the depth of the 2007–09 Great Recession. But it took only a couple of years for economic activity to get back to pre-recession highs, before the Roaring Twenties got underway in earnest. Unfortunately, more recent recoveries have been much slower.

So a danger highlighted by the 1918–19 recession is that, just because a recession ends and a pandemic passes, it doesn’t mean blue skies ahead. A more recent example is the eight-month 2001 recession that ended in November 2001. During that downturn, the S&P 500 plunged to a new low the week after the 9/11 attacks, but then surged more than 20% in 3½ months. However, stock prices turned down in early January 2002 – ahead of a new growth rate cycle downturn, i.e., a mere slowdown – and fell to a much lower low by October 2002. Those historical precedents behoove us to keep a close eye on our leading indexes.

Kotok: When you wrote your forecast two months ago, you had the best information available then. Things have dramatically changed worldwide. COVID-19 cases and deaths, US-China tensions, and the Hong Kong financial center alteration are just some examples. How do you make adaptations for events like these, and are they considered extreme adjustments?

ECRI: When we wrote that piece in early April, we explained that a recession’s severity was measured by its depth, diffusion and duration – the 3Ds – and explained why the recession would be “extremely deep, very broad, but relatively brief.”

At the time, ECRI’s publicly available Weekly Leading Index (WLI) was “in free fall,” but we expected some answers from the WLI and our other leading indexes regarding the evolving future course of the economy.

Today, with several weeks of additional data available, we see our expectation of a “relatively brief” recession bolstered by sequential improvements in our leading indexes, including nine straight weeks of increases in the WLI. This constitutes objective evidence that our earlier thesis had been correct: The recession is likely to end by summertime. In fact, because stock prices always turn up a few months before business cycle troughs, their upturn is also entirely consistent with that prospect.

We also wrote: “After we restart the economy’s engines, it will pull out of its nosedive. But without all engines firing, it won’t be a steep climb back up.” Rather, we expected “a slow, halting recovery from a severe economic contraction.” On this score, our assessment hasn’t changed.

Kotok: ECRI has an enormous and prestigious reputation in its approach to the business cycle. The elements we have discussed are not part of the normal cycle. Each of them is in the category of “shocks.” It seems like we are in living through a continuous series of shocks. Doesn’t this force a rethinking of the traditional cycle forecast models?

ECRI: History is punctuated by shocks, large and small. But because we have a good understanding about what is cyclical, it makes it easier for us to strip out the cyclical component to reveal underlying structural developments sooner than most. For example, we warned of the structural decline in trend growth in the summer of 2008, more than five years before “secular stagnation” became a hot topic.

And recall that the unrest of 1968 happened to be followed by the mild 1969–70 recession. Yet, the more consequential structural change that followed that period of unrest and uncertainty was the downshift in productivity growth in the early 1970s. Today, we are concerned that the uncertainty created by the unrest will inhibit investment, further curbing the economy’s productive capacity.

What our late mentor, Geoffrey H. Moore, called business cycles “of experience,” i.e., as actually observed, were always influenced not only by exogenous shocks, but also by endogenous cyclical dynamics, meaning that a complete explanation of any historical business cycle can’t be purely exogenous or purely endogenous. It’s really about both endogenous cycles and external shocks.

For example, some believe that the San Francisco earthquake was the shock that triggered the chain of events that culminated in the panic of 1907. Yet, what’s notable to us is “the dog that didn’t bark”: Though the earthquake happened in April 1906, no recession occurred that year. From our perspective, despite a major external shock, there was no recession precisely because of the endogenous cyclical dynamics of the U.S. economy that were not predisposed to one. We know this because, at the time the San Francisco earthquake hit in 1906, our leading index was in a clear cyclical upswing, signaling that the endogenous cyclical forces would ensure the economy’s resilience to exogenous shocks. The precise nature of the shock didn’t really matter. But by early 1907, the configuration of the endogenous cyclical forces had changed, opening up a recessionary window of vulnerability, as signaled by a cyclical downturn in that leading index. The 1907–08 recession started soon thereafter.

The December 1941 attack on Pearl Harbor was a huge shock, but didn’t trigger a recession before wartime deficit spending had a chance to boost economic activity. The reason was that the economy was in a cyclically resilient state, as shown by the strong upturns in ECRI’s leading indexes in the lead-up to that attack.

You see, a good leading index of recession and recovery should be conceptually sound, and have the ability to capture the endogenous cyclical forces that determine the economy’s resilience to exogenous shocks. When such a leading index enters a cyclical downturn, it signals the economy’s entry into a recessionary “window of vulnerability,” within which any significant exogenous shock will tip the economy into recession. This is key to assessing recession risk on an ongoing basis.

Such well-designed leading indexes work in other market-oriented economies, as well. Readers may recall that the consensus view going into the June 2016 Brexit vote was that it would trigger an immediate recession. But the week beforehand, based on our U.K. Long Leading Index, ECRI reaffirmed that a vote in favor of Brexit wouldn’t be recessionary. This was because our Long Leading Index was in a cyclical upswing, signaling a U.K. economy resilient to exogenous shocks.

In other words, in our analytical framework, what matters is whether the endogenous cyclical forces measured by our leading indexes are so arrayed as to open up a recessionary window of vulnerability, within which any of those kinds of shocks would trigger a recession. And so we’ll continue to focus on our leading indexes for primary guidance about the cyclical risks to the economy.

Kotok: Last question. If you had to write or rewrite the outlook piece of two months ago today, what would you do differently? What would the outcome be, knowing what you now know but had to estimate two months ago?

ECRI: The main conclusion of our piece in early April – “that this recession will be extremely deep, very broad, but relatively brief” – remains valid, and would not benefit from hindsight. In fact, the NBER Business Cycle Dating Committee yesterday reiterated our central point about the three D’s, writing that, “in deciding whether to identify a recession, the committee weighs the depth of the contraction, its duration, and whether economic activity declined broadly across the economy (the diffusion of the downturn).” In that regard, the Committee wrote that “the unprecedented magnitude of the decline in employment and production, and its broad reach across the entire economy, warrants the designation of this episode as a recession, even if it turns out to be briefer than earlier contractions.”

With regard to the nature of the recovery, in early April we posited that “the public may not be eager to resume going to restaurants and entertainment venues until they feel safe. That could take a while, and will hamper the services side of the economy. They’ll venture out only when health officials can credibly assure them that the pandemic has been contained, and that a vaccine or, at least, a therapeutic drug – is available. Indeed, while we all appreciate the urgency of reopening the economy, saving lives is a precondition for saving livelihoods.”

However, with the benefit of hindsight, we now know that many will not abstain from going out to restaurants and entertainment venues until “health officials can credibly assure them that the pandemic has been contained, and that a vaccine or … a therapeutic drug is available.” The mood has changed, and it’s now clear that many prefer to prioritize normalcy, their livelihoods, and protests over any perceived risk to their lives.


Closing Remarks (Kotok)

Whether the recovery is a V or a U or something else is still an open question. But highly credible forecasting work will help us determine the direction of the US economy and, because of the ties to that economy, the US stock market. Global influences are now huge, and international policy choices will assist or impair those effects.

Bottom line: We may have just seen the sharpest and fastest decline in history, which bottomed in April. It is the way back that counts now. How fast and how far remains to be seen, and credible forecasts help frame the questions.

We are most appreciative of Lakshman and Anirvan for sharing their views with our readers.

Review ECRI’s real-time track record.

For information on ECRI professional services please contact us.

Follow @businesscycle on Twitter and ECRI on LinkedIn.

Related News & Events

Biggest Risk for Markets

Yahoo June 24, 2020

Interview on recession severity and the biggest risk for markets. More


Yes, We're in a Recession. But the Market's Rally Still Makes Perfect Sense

CNN June 24, 2020

With the economy officially in the deepest recession in living memory, the stock market's ongoing rally can seem baffling. More


Why a Ruinous Recession and Soaring Stocks Make Perfect Sense

Investopedia June 13, 2020

A rising equity market historically leads an upturn in the business cycle. More