Friday, March 26, 2010

What a Real Recovery Looks Like: and Why We Likely Won't Have One

I am going to examine (graphically) the recovery follow the early 80's recession(s). Specifically, I am going to look at the recovery following the very deep 7/81-11/82 recession that saw unemployment peak at 10.8% and had a major decline in GDP (like our current recession). I am also going to show a comparison of where we are now (compared to where we were then) using the St. Louis Fed's assumption that the recession ended in July 2009 (meaning 7 months from the end). Finally, I am going to demonstrate why we had a very robust recovery following the end of the 1981 recession and why we are unlikely to have a robust recovery this time.

So, let's take a journey back to the early 80's when The Smith's were just getting started, the Weather Channel went live, Ripken had just began his streak, and the last decade in which the Yankees didn't win the World Series.

The recession that began in July 1981 (and followed on the heels of a recession that had just ended a year earlier) was the deepest (at the time) since the depression with unemployment topping out at 10.8% and GDP falling 2.6% over the recession. The recovery that followed this recession, however, was very robust (for several reasons) as we can see here looking at the decline in the unemployment rate (compared to GDP) and the change in payrolls (I adjusted this to compare to today's much higher population in rough terms):
gdpvunemp

adjusted payrolls

As you can see here, the initial recovery was incredibly strong in terms of both employment and GDP, but as the decade languished, unemployment remained stuck around 7% even with consistent GDP growth above 2.5%. The reason I used an adjusted payroll growth number was to give a better illustration for purposes of comparing the positive job gains of the 80's recovery to what we may begin to see later this year (likely next month).

Now, let's look at some comparisons to where we are now versus then (pardon the hard to read nature of these):

payrolls comp

This first graph shows (in percent change terms) payroll growth coming out of each recession. Not only do we start with a larger hole to dig out of, but the early 80's recovery was already positive by now and then exploded over the course of that first year or so (we can use this graph to compare later on this year).

Next, looking at GDP shows us that we are already behind the 80's recovery here and that it is looking highly unlikely that we are going to catch up
gdp comp
as the initial 80's recovery saw two quarters in excess of 7.5% to kick it off, while we are looking at two Q's just above 5%. That will come back to haunt us later.

Finally, let's look at some of the drivers of the 80's recovery and see why we are not set up nearly as well as back then. First up is the elephant in the room, interest rates:

rates

While these look very high compared to today, they essentially fueled the recovery by allowing significant refinancing (and lower business costs) activity that continued for the next 25+ years to some extent. The level of these rates also kept the economy more honest, as everything had to be evaluated against a 10-year bond that was never below 7% for the decade (ie no need to chase yields and speculation). These interest rates also kept savings high

savings

which help to provide an economic backstop and can fuel growth as the rate declines (not necessarily good in the long run, but can lead to larger growth over a shorter term). Finally, we need to look at debt levels, as expanding levels of debt (to a certain point) can have a huge impact on growth

debt percent

and as we can see here, as the 80's progressed we saw debt service jump by nearly 20%, which helped to fuel the expansion.

So what does all of this say about the situation facing us today? Well, we do not have the catalysts that the 80's had to fuel a large economic expansion. Our savings rate is low and looks to be rising (good for long term, but bad in the short run). Our debt service rate high (and even after a spectacular decline is still higher than the 80's peak), which will limit our ability to use household debt (ie consumption) to jumpstart a recovery. Interest rates are already at historic lows, which will prevent significant refinancing activity from helping out families and will also create a huge ceiling on home prices going forward as rates rise. Finally, our initial burst out the gate simply wasn't fast enough to make up the ground we lost. In other words, we are likely going to be facing a recovery/expansion that has many more risks to the downside than it does catalysts to the upside and we will be doing it this time without the luxury of having a Keynesian backstop should we falter again.

For Bruce:

savingsv10year

5 comments:

Anonymous said...

There are a lot of good points to be made there but at the very least we are doing better than a lot of people expected a year ago and we managed to escape hitting or exceeding the peak in unemployment of 1983.

BruceMcF said...

Note that high interest rates on their own cannot raise savings and low interest rates on their own cannot lower savings ... net new savings have to be consistent with the financial assets where savings can come to a rest ...

(1) real investment in newly produced productive equipment and construction, plus

(2) Government deficits, plus

(3) Current account surpluses

Our low savings rate is due to our much higher structural trade deficit than we had in the early 1980's, and the much lower rate of investment in productive equipment than in the early 1980's.

SilverOz said...

Oh, I don't know. i think the personal savings rate is fairly well correlated with (and i understand this doesn't mean causation) interest rates (and it makes perfect sense too). See the new graph at the end of the post.

George Phillies said...

I am looking at your real GDP and Payems graphs. They are showing sharply climbing lines with no indication yet of a plateau. Other indicators are doing the same thing. Until those lines roll over, it appears to bit early to say that we have a slow-recovering recession rather than a long-lived recession.

brodero said...

One of the problems with the personal savings rate is that it is
residual (income minus outlays = savings)....it does not account for the deleveraging that is going on....