Business Cycles: The Recession Dating Game
It is interesting to look at the dates of recessions and expansions and frame them against the dates when the NBER decided.
As I start publishing again on market topics, I am reentering the fray with a series of multicycle time pieces that set the table for weighing the direction of this convoluted and much-debated cycle. As we gear up for a very rough road ahead, these types of “lookbacks” can be of use for background. Today I am back on the recession topic.
Are we in a recession? Who gets to decide?
With all the recession debates going on since the spring, we thought a quick look at “official” recession periods according to the NBER’s Business Cycle Dating Committee was worth detailing. This will at least give some context to all the noise. In the chart below, we include the dates when the Committee announced the beginning and end of each recession. Note the lag periods.
At present and as expected, there is significant political pressure from the GOP to hammer home the concept of “recession” into the voter psyche. With midterm elections a few weeks away, the two consecutive quarters of GDP contraction in the first half of the year set off an enthusiastic scramble by those not in power to make the recession tag stick. They can make a defensible case, but we think it falls short and the jury needs more data points in coming months. On the recession chatter broadly, politicians have opinions and sometimes even facts. But they don’t make the call. They are more than axed and bring an incomplete case. They certainly seldom discuss employment’s positive factors, and that undermines credibility.
Politicians weighing in on economic performance in troubled times is always a challenge. To this point, can you imagine the executive branch being the arbiter of its own recession/expansion labels? You wouldn’t catch Biden calling this a recession even if there was a soup line that ran from the Washington Monument to Union Station. Fortunately for Biden, soup lines are not the main event at 3.5% unemployment, a rate that stands as a low since the start of the 1970s. You have to go back to the 1950s and 1960s to find only a few lower. That does not change the fact that he drew the short straw on inflation fallout – some totally unrelated to his actions and some quite related. He took a short straw and made it shorter.
Meanwhile, Trump would say that he – and he alone – had created the greatest economic expansion since the dawn of civilization even though he never got north of a 2% handle annual GDP. He had only four 3% handle quarters offset by three 1% handle quarters and one sub-1% quarter. If we include the 2020 COVID numbers, Trump had a materially lower average annual GDP than Jimmy Carter (who is often a one term whipping post in political discussions). Trump posted the lowest annual average GDP growth from the Carter to Biden years. Even if we’re generous and leave out 2020, Trump’s average GDP growth still falls short of Carter. And Trump is not alone. Obama came into office with a systemic crisis being dealt with, but he also failed to clear the bar set by Carter on average GDP growth. The Obama to Trump years never cracked a 3% handle annual GDP – not even once.
In fairness to both, the slow growth US economy is a very complex set of issues beyond the person in the White House. The bottom line is that small numbers make big differences in low growth economies. This is the challenge in framing recessions and weak economies these days, and as always, the devil is in the details. Trump’s numbers were simply low, but Obama’s were volatile with four negative quarters offset by three 4% quarters and one 5% quarter (in 2014). Those eights years of Obama stayed in the 2% range overall – just like Trump. To ignore the fixed investment aftereffects of the credit crisis from housing to employment and the structural changes that came in the crisis aftermath is just dumb or political. The same with COVID. Research needs to be politically agnostic and objective – even if you can have some fun with it. The term snowflake has come into use in recent years, and there are a lot of them out there in each political party. They often get touchy in the face of a colorful or liberal use of economic facts.
Does the Fed determine recessions?
For obvious reasons, the Fed (led by an exec branch appointee even though the President can’t fire him) does not make the “official” call on recessions either, even though their dual mandate of price stability and full employment (essentially inflation down/jobs up) requires them to act (or not) based on recession concerns. However, it is not their job to avoid recessions at the expense of high inflation. They just don’t say that too frequently with their outside voice. Their job is to focus on financial stability, prices, and employment, and they don’t weigh in on fiscal matters or UST purviews such as the dollar. As discussed in my recent piece on the Misery Index, the Fed’s job is not as easy as the mandate might suggest as inflation and job metrics often march to the beat of different drummers. This is the case right now with stagflation-esque numbers running counter to payrolls and consumer demand across some major industries e.g. autos and leisure, which tell a more mixed story.
Unsurprisingly, my view is that the market and its participants decide when there is a recession based on where and how they invest. Though not “official,” it is real and in real time. Corporate managers also decide by way of hiring and capital budgeting. For the market players, that comes with the asterisk that they can change their minds as often as they like and can have investment time horizons using views that vary widely.
NBER’s Business Cycle Ratings Committee gets risk free overtime…
One of the main points of this note is to provide past recession details while presenting some of the strange timing twists around NBER’s “official” recession start dates. The chart shows the time it takes for the NBER’s committee of eight sitting dignitaries to weigh in. We put quotations around “official” since NBER is a nonpartisan independent entity and not a government agency. It does get grants from the government and private sources. At their pace, we assumed they get paid by the hour (they don’t of course). Established in 1978, it is generally recognized as the leader in tracking cycles with top shelf research. It is basically an economic brain trust of top dogs in the field. Some members have plenty of time served in Washington in important roles.
In the chart, we line up recessions from most recent to distant since the committee’s inception in 1978, i.e. the first recession they delineate is from Jan 1980 to Jul 1980. We show the dates of the recession and its duration in months. As a reminder, the longest recession since the Great Depression stands at 18 months from the Dec 2007 peak to the Jun 2009 trough. That recession passed the former tie for #1 of 16 months held by the 1973-1975 recession (not shown) and the Jul 1981 to Nov 1982 recession. The shortest recession ever is on this list with the 2-month recession from the Mar 2020 peak to Apr 2020 trough. As we mentioned in our comments on how unemployment factors into recession arguments, we have a hard time blowing the whistle and calling a recession now with the unemployment rate at 50-year lows (see Unemployment, Recessions and the Potter Stewart Rule). We see only a couple of years from 1950 to 1970 lower.
Key Takeaways
Bottom line on NBER and the Faculty Club lag times…
The delays are easy to make fun of since most people want quick answers and objective recipes to use (e.g., two quarters of contraction). However, line items are not all created equal and that’s an important factor. When growth rates are around a 2% handle on average – as they were under both Trump and Obama – one quarter of negative net export numbers, some inventory liquidation, or some government budgets cuts can send you into the negative zone even with positive PCE. The lag times are almost silly in this process. Sometimes you feel like they have gathered up a group of dignitaries from the Baseball Hall of Fame and put them together in the fall to tell you who won last year’s World Series. After giving so many students and doctoral candidates deadlines, you would think they would speed up the pace.
Why the lags dilute the value of the recession designation process…
My personal favorite was the decision in Dec 2008 to identify Dec 2007 as the start of the recession. That Dec 2008 decision was made as the systemic crisis was mushrooming across the financial system, Lehman had failed, Merrill was quickly folded into BofA, the FHFA had taken over Freddie/Fannie, and the government owned AIG. What was the committee’s first clue?
Even by the Dec 2007 retroactive recession start date, the credit markets were in disarray, RMBS was an unfolding horror show, the “hearts” of more than a few hedge fund managers had been crushed in a garlic press, illiquid collateral was setting off internal risk panics across derivative and structured credit counterparties, and bellwether names like Countrywide were getting de facto bailouts in the private sector (Countrywide was in the process of being recapped and then merged into BofA). The year 2007 posted four quarters of positive growth, so the recession had not arrived – yet. The fuse was lit, and the fuse was short.
On the heels of Bear being rolled into JPM in March 2008, we saw very rare negative PCE lines in 1Q08 and 3Q08 and again in 4Q08 with three of four quarters posting double-digit declines on the Durables Goods line across 2008. Further, residential investment was in a freefall and had been negative since 4Q05 with double-digit declines every quarter from 2Q06 all the way through 2Q09. The list goes on from there. The credit market was toast in the summer of 2007. Equities peaked in the fall. I’ll go out on a limb to suggest that NBER could have made a recession call a bit earlier.
Does the official “recession” term even matter?...
In short, YES. That recession designation does matter to many since it influences the report cards for elections and appointed officials. The popular use of the term is engraved in our hearts and minds as a symptom of economic pain. The designation of “recession” is used to grind axes against officials as each side struggles to win and therefore gain support for their policies (and judicial appointees). The “recession” tag (arguably) hurt both Carter and GHW Bush. The end of the recession for Bush I was designated as Mar 1991 by NBER in Dec 1992 - just after the election. The report of a return to expansion could have helped – “it’s the economy stupid” as the slight misquote of the Clinton team is phrased. Bush could have used the independent stamp on the “fact” that the recession ended 20 months before the election. At the very least, he could have been able to shut Ross Perot up (I doubt it though). In the end, Clinton became President with 43% of the popular vote.
The timing and voter reactions to “recession calls” can have pivotal impacts on policies that drive the direction of trade, taxes, financial market regulation, energy policy (e.g., pipelines/infrastructure, drilling incentives, climate legislation, etc.) and the rest of the laundry list. That’s pretty damned important.
In contrast, the timing of “recession” dates has very little to do with how securities swing around since the designation is very much a lagging indicator (like the default rate) that comes after metrics deterioration has already been absorbed by the market. We would argue the official start of the recession is not a late breaking news item to the person who has been unemployed for a long time (Worker to Cycle Dating Committee: Thanks for the news flash, Professor, my unemployment benefits ran out already and my health care is gone. Appreciate the post though).
What matters in markets is when the owner of securities or lenders decide the recession is here. That is often a battle between bulls and bears that gets fought out among those with varied investment time horizons from years to months to days to lunchtime. While the recession label needs some empirical support, the committee’s final recession crowning label tends not to add much value with institutional investors.
Navigating the GDP line items: The 1Q22 mix of line items as listed in the GDP accounts was not as worrisome for most investors as this year proceeded. The bigger focus was on rising interest rates and inflation. Handicapping the effects ahead for a UST shift was more important than a -3.1% swing in GDP contribution from net exports in 1Q22 that flowed into the -1.6% GDP line. The sustained inflation, the expected fed funds rate increases, and potential yield curve fallout were the bigger catalysts in the markets. The material contraction in investment and slower PCE Goods (vs. Services) growth in 2Q22 was more important in a smaller negative headline GDP contraction of -0.6% in 2Q22 vs. the -1.6% in 1Q22. The 3Q22 earnings season will bring more color on how much fixed investment dropped at some bellwethers and how that could continue into 4Q22.
Obviously, the priority indicators go well beyond the GDP headlines in framing a recession. For example, Trump saw a sharp drop in Gross Private Domestic Investment in 4Q19 at -6.5% before COVID made that all irrelevant. Biden just saw a double-digit decline of -14% on that same line vs. the preceding quarter as Residential was hammered by -17.8% and Nonresidential Structures by -12.7%. Equipment ticked down slightly while IP investment stayed strong. Naturally, these investment lines are a worry for the direction of the economy now. More negatives there and weaker PCE or softening consumer signals could be a trigger for more selloffs and a move closer to a recession consensus in the market. We are not there yet. The employment data last week also made a relatively constructive statement on fundamentals. Jobs matter.
Swings in Personal Consumption Expenditures (PCE) is the main event: The headline growth rates in GDP are important, but under the headline level, I see PCE as the top focus at around 68% of GDP (2Q22). The moves in Gross Private Domestic Investment are critical ingredients also given how they can swing into sharply negative numbers. It is hard to have a true recession with positive (even if minimal) PCE growth and very low unemployment. The spikes in inflation and strain on real wages are flashing red for now as we follow how it plays out in household discretionary cash flow, potential pressure on home valuations, an inevitable stall in new hiring rates, and possible trimming of current payrolls in select industries to support profitability.
PCE drives the bus, but Gross Private Domestic Investment (GPDI) supplies the gasoline: GPDI is the GDP line item that is more like capex in issuer and industry analysis. Corporate sector spending brings in those old-fashioned economic terms like “multiplier effects” (household spending on the back of job growth, higher supplier-to-consumer chain activity, support for freight and logistics services across the chain, household and corporate financial service needs from working capital financing to insurance, and growth in the tax base from state and local to federal, etc.). One big risk to the business cycle now is that the GPDI line is in the crosshairs for trouble. We just saw the worst GPDI quarter in 2Q22 GDP (-14.1% from preceding quarter) since the COVID quarter of 2Q20. The GPDI line put up a -2.8% GDP contribution to a GDP headline number of -0.6%. Before the 2Q20 COVID quarter, you had to go back to the 2Q09 cyclical trough quarter to find a worse performance in GPDI than 2Q22. The 1Q22 GDP headline contraction line of -1.6% was distorted by a -3.1% line in net export contribution. The weakness in fixed investment in 2Q22 brings a much stronger warning than the 1Q22 numbers. Nonresidential Structures and Residential Investment were the big losers in fixed investment. Capex color in 3Q22 earnings season and any material changes in plans will be a very important focal point.