It is fair to say the odds of a double dip recession have lessened slightly in the last week. The change in the data have tilted that way, without actually altering the moderate downward slope of the trend that has been in place for about two months.
The ECRI leading Index, WLI, may show another slight increase on Friday, but the more important ECRI Growth Index will still likely drop further, even if only 0.1%.
This post though is not about individual data points.
It is about the concept of the double dip, what it means, and if we are going to have one this time. My conclusion: the term double dip is not useful, the real question is whether it is two separate recessions or just an extension of one.
Let’s go back to 1980-1982, the last and only “double dip” recession. Before the recession began in 1980 the economy had peaked in Q1 of 1980 at $5.908 trillion in constant dollars. From there the economy headed downward to a trough in Q3 of 1980 at $5.776 trillion. That recession ended in Q1 of 1981 when the economy increased to a GDP of $6.005 trillion.
There is no hard and fast rule for when a recession begins and ends. That’s why the NBER has a dating committee. Incidentally the recession dating committee has not yet ruled on an ending date for this recession, a possible reason for believing it has in fact not yet ended.
One common sense way of dating a recession’s end is when the economy comes out of its trough and rises above the previous peak, as in 1980. So in theory if another recession were to follow right after that one, it would be a separate recession and not part of the first, since the first one had already ended. The recessions in 1980 and ’82 were two different recessions, though some argue with some support it was really one. A somewhat similar but different situation occurred in 1937, when the recession within the Depression occurred long before the peak was reached, but long after the bottom had been passed. Everyone agrees though it was just one long event. In that case the Depression had not reached its previous peak before 1937.
This is not just a question of semantics or a hypothetical question. Analyzing two smaller recessions or just one larger recession could lead to different conclusions.
The possibility of a further economic decline from here is very real. So far the economy has dropped 4.1% from $13.36 trillion to $12.81 trillion, making it the deepest recession of the post war era. The economy has come back, but is still 1% short of reaching the previous peak. It is doubtful it will reach that level in Q3 and it is possible the economy can head down again as early as Q4 thus preventing it from reaching its prior peak. That is not a forecast, just a very real possibility.
Another downturn starting in Q4 or possibly in Q1 would therefore not be a different recession, but just an extended part of the original recession that started just over two years ago, using the definition from above. That’s actually a worse situation than a second recession because you have to have a pretty large problem before you can have one decline being a part of another.
In 1981 after the recession ended in Q1 the economy remained strong for three quarters at roughly the same level before heading down again. There were 4 quarters of real growth, with three of them above the previous high. This was more than enough time for companies and consumers to reset psychologically, financially, and in their expectations. The reset is what functionally made it two recessions back to back and not one long recession.
Unlike 1980-1982 where the first episode was mild, in the current case the first episode was major. In this case businesses and consumers have not had nearly enough time to reset with over two years of recession imprinting since Q2 2008. Evidence for this is both anecdotal and in the fact we have not yet reached the previous peak. The magnitude of the insult was large enough to indelibly impress caution on all actors for a much longer period in the context of an improving economy, while fear remains of another downturn.
As has been pointed out by others, given the slow expected growth trajectory, the chances of another below zero growth episode is now much greater, just because it is easier to fall below zero when starting at 1% growth rather than 4%. Just as important, it is more likely to have another downward episode because all the actors still have not recovered from the first episode altering their previous normal propensity to spend. As they are just in the process of a long recovery they are likely to react with unusual caution to positive news and over react to any negative news in their industries and in the macro environment. This all loads the dice in favor of a new downward episode.