Posted by: Josh Lehner | January 26, 2017

Expansions Don’t Die of Old Age

In a recent meeting someone mentioned that much of the current state budget discussion is based on structural issues — revenue growth not keeping up with expenditures — but at some point in the not-too-distant future we will be back to having cyclical budget issues. In our world that counts as gallows humor and we all nodded our heads in agreement because it is undoubtedly true. Such issues are what keeps forecasters up at night. The comment also dovetails with a question we hear a lot in presentations, something like “Aren’t we due for another recession?” Well, the short answer is no. The long answer is yes another recession will come, but it won’t be because we hit some magical length of time or duration of expansion.

That said, the current economic expansion is getting a little long in the tooth based on historical patterns. Next week we will tie the 1980s expansion for the 3rd longest since WWII. Many economists today are now talking about a maturing business cycle. Here in Oregon, as we approach full employment, we’re seeing job gains slow. Nationwide it appears we’re now somewhere past the peak of the cycle in terms of growth rates. Now, the expansion is expected to endure for awhile longer — the WSJ consensus puts the probability of recession at 16% — but it’s obviously clear we are no longer in the early stages of a recovery.

In fact the current expansion, at least in terms of length or duration, is beginning to enter rarefied air. The last time the U.S. economy was in expansion this long was back in October 1998. Aerosmith’s “I Don’t Want to Miss a Thing”, Barenaked Ladies’ “One Week” and Tim McGraw’s “Where the Green Grass Grows” were topping the charts. And the time before that was in the summer of 1990 when Madonna’s “Vogue” and New Kids on the Block’s “Step by Step” went to #1. So, yeah, it’s been awhile.


Thankfully there is no real mathematical relationship between the length of an expansion and the probability of falling into recession. Now, there is some relationship, but it is weak. As an expansion hits 7, 8, 9 years in duration the probability of remaining in expansion still tops 90%. That said, this simple model based on actual U.S. experience necessarily blows up at 10 years which is the longest expansion on record. It’s not that an expansion can’t last longer than that, just that we haven’t seen it. Furthermore, when it comes to Oregon state revenues, we went nearly 20 years without a major revenue event, from the early 1980s up to the 2001 recession. The reason was the 1990 recession was mild and while revenues stopped growing, more or less, they did not plunge like they typically do.


Ok, so why do recessions happen? Well, they can come in various forms — supply side shocks, financial crises, currency crises and so forth. However the story economists usually tell is about imbalances. These can be sector-specific — high-tech in the 1990s or housing in the 2000s — but usually results in a generally overheating economy. Price pressures build as demand outstrips supply. The economy operates for a time beyond full capacity. As such inflation rises and the Federal Reserve steps in to take the punch bowl away. This was at least the typical pattern prior to the last few recessions.

Former state economist Tom Potiowsky likes to say that expansions don’t die of old age, they die due to mistakes. Those can certainly be policy mistakes, monetary or fiscal. However there still generally needs to be some sort of coordinating event to turn any growing pessimism into a recession. Some sort of Wile E. Coyote moment when it becomes clear we did run off the edge of the cliff. Then a recession ensues and we start the cycle again.

Where exactly the economy is in this chain of events is a hotly debated topic. While I don’t need to tell you this, but I will anyway, the economics profession hasn’t exactly wrapped itself in glory in accurately forecasting recessions in advance. As one of our advisors says, however, we’re really good at autopsies. Yet we still try our best to gauge where the economy is, how it is performing relative to the Fed’s dual mandate and the like. As stated above there is typically a period of time, somewhere between 2 and 4 years, where the economy operates beyond full capacity prior to the next recession. Today, the U.S. economy and here in Oregon is just now approaching or reaching many estimates of full employment. Using the past as a guide, there is likely still room to run for this expansion.


Unfortunately there is no silver bullet for forecast uncertainty. Our office does at least three things to help provide context for the outlook. First, we use our leading indicators and Tim Duy’s as well to help us with turning points in the economy. Such indices can be best thought of as green light/red light indicators. Second, we provide historical revenue forecast errors along with revenue tracking each quarter, which help show how much uncertainty remains as the biennium progresses. Third, we also provide a set of alternative scenarios — developed with the Legislative Revenue Office and based on historical examples — to try and show what sort of economic and revenue changes could be expected should a recession begin in the next year or two. That said, these items obviously do not solve the budgetary consequences of an actual recession should it occur.


  1. […] Source: Expansions Don’t Die of Old Age | Oregon Office of Economic Analysis […]

  2. Josh- Great article. Very easy to read and understand. Lot of good info. We have a couple talks coming up, so the timing for us is very relevant. Best regards, and Happy New Year (from Vietnam)- Mark

    Sent from my iPhone


  3. […] Expansions don’t die of old age | Josh […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: