Courtesy of Doug Short.
Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
More than a year ago, in spring of 2013, I wrote two posts about how the retirement wave of the Baby Boomer generation would result in much lower growth, but full employment (see US Economy: Below Stall Speed or Already Above Potential? and Forget the Jobless Recovery, Get Ready for the Full-Employed Recession). Though more and more economists now recognize that it is to a large extend this retirement wave that has led to a huge exit out of the labor force, measured by the falling labor force participation rate, few fully anticipate the wide spread implications of this development. Here is the probably most crucial of all the charts I posted, updated with last year’s number and an estimate for this year: The relationship of economic growth and the change in the unemployment rate.
The solid line is the regression between these 2 parameters. Historically the unemployment rate has held steady when GDP growth was around 3% (intercept of the regression line with the x-axis). As the labor force is the most crucial limited input factor in the economy, the growth rate at which unemployment is stable must be the long-term potential growth. Now what happened since the first cohort of Baby Boomers, the generation born from 1946-1964, reached their usual retirement age of 65 in 2011? Unemployment fell much faster than GDP would have predicted. And this picture has prevailed ever since. The dots for 2011, 2012, 2013 and the forecasted one for 2014 (2014e=expected, based on the current Bloomberg GDP estimate and annualized year-to-date change in unemployment) lie all further away from the historical regression line than any year before, implying that the observation I made last year was no fluke, but rather a new trend. Now the troubling implication is this: If you assume that the slope of the curve is unchanged and would shift the solid line up to the years 2011-14 you would get the totted line that would implicate potential growth of about zero!
Now, though not the entire fall in the labor force participation rate (green line) is due to the retirement wave and might be reversed, a very large part is (orange line is the participation rate of prime-aged workers, which is not effected by retirement).
So while there is indeed slack in the prime-age working cohorts, this implies that the overall problem is more difficult to reverse than generally assumed. A look at each cohort’s size of the US population shows further trouble ahead.
Based on data from the US census website for 2013. (http://www.census.gov/popclock/)
Now first you can see the jump due to the 1946 onward baby boom. While this cohort has already moved in the retirement age, the trouble that lies ahead is the still sharp rising number of retirees in coming years (red arrow on the right side), while at the same time the number of people entering the labor force is falling (red arrow left). The situation is about to worsen till about 2025. Given that there are nowadays more mid 50s than teenagers for each cohort that means the labor force is bound to shrink, if not for immigration (which currently is a hot political issue), while the population as a whole will further increase, resulting in a worsening dependency ratio.
That sounds quite gloomy, but what are the actual implications? My main issue in my posts last year was that while everyone was complaining about low growth, growth was actually much faster than you could long-term expect (that is until about 2025). While people talk a lot about a “new normal” and how the Great Recession has slowed credit growth and therefore economic growth, my opinion is that only the slack from the recession made the current growth possible. Now that unemployment is fast approaching what economist regard as “full employment” (or the “natural rate of unemployment”, which is estimated by the Congressional Budget Office), labor shortages are becoming wide spread and growth will actually slow further till it reaches a new equilibrium, where unemployment is steady.
This was what I thought about when I last year wrote about temporary full-employed recessions, something completely unusual to the US, but not that unusual in other countries that have a shrinking work force. Growth will be so slow that the economy will just due to the usual short-term fluctuations temporary shrink without causing the usual hallmarks of recessions like a spiking unemployment rate. The 1st quarter of this year, though weather-related, may be sign of things to come.
So this means low growth, but also low unemployment.
What about inflation?
Low unemployment (or rather the difference to the natural rate) is usually a sign of coming wage growth.
Now while this might be the case now, as more and more indicators signal rising wages, the implication for inflation is much less clear than you might think.
Honestly, I for some periods like the entire 1990s I don’t see any meaningful connection.
More important, an aging population actually puts downward pressure on inflation.
(The following 2 charts can be found throughout the web on a couple of sites, though I didn’t find their true origin. I took them from a post from John Mauldin that was rather centered about Japan.
Why? Because an aging population takes out less credit and credit is so much connected to inflation.
Young people take out credit when they enter the labor force, buy a house, a car, start a family. You don’t take out a loan, when you retire; you may not even get one.
For that reason, the inflation-wave in the 1970s could actually be attributed at last to some extend to the exploding labor-force that occurred as the Baby Boomer generation started to work.
Interesting how this relationship between workforce growth and inflation is even much closer in Japan:
Unfortunately the above population time-series that I took from FRED ended in late 2013. But the last part of the chart is particularly interesting. Japan’s post-war baby-boom was much shorter than in the US and the usual retirement age there was until recently 60, not 65 as in the US. As a result, Japan’s retirement wave is actually over now. And that may have an impact at least as strong on the recently rising inflation (and therefore the long-awaited end of deflation) as the much hailed Abenomics.
So that means we could now see actually a mirror image of the 70s: Low unemployment and low inflation. That may mean that the unusually (and probably unsustainable) high US corporate profit margins may erode, as wages rise, but can’t be passed on in the overall disinflationary environment. But it could also mean that companies are less willing to raise wages when faced with labor shortages than in the past. “Good” well-paying jobs might still be scarce despite a shortage of labor. At least some anecdotal evidence point also to the latter. Or it might be a mix of both.
So low growth, low unemployment and low inflation (until around 2025). Where does that leave the Federal Reserve and the financial markets? With inflationary pressure absent, there is less need to raise rates, but with unemployment lower there is also no justification to conduct quantitative easing (QE) forever and let the FED balance sheet explode. So while a see QE to end in October as planned, I don’t see any meaningful interest rate hikes either. For quite some time people have feared that the more than 30-year old bull market in bonds will end in a big crash, once the Fed starts raising rates. I don’t see that.
Have a look at the Fed’s last 3 big tightening cycles and what they meant for bonds:
While they led to some kind of crash in 1994 and 1999, the long end of the yield curve completely ignored the last one, though it was in terms of change in the Fed Funds Rate the by far most severe, which made Greenspan at that time scratching his head and speaking of a “conundrum”, and a “savings glut”, though there might be a much simpler explanation. The big difference was the slope of the yield curve in 1998 and 2004. In 1998 it was flat, even partly inverted, while in 2004 it was very steep. That let the market absorb a huge hike in the Fed Funds without a change at the long end of the curve. 1994 was a bit different. The curve was rather steep, but too many were at that time overexposed to US Treasuries. The yield curve carry trade was the most popular thing to do. When Greenspan unexpectedly raised rates in 1994, too many found themselves at the wrong end of the trade and had to cut their losses (like famously Orange County, at that time the most wealthy district I the US, which went bankrupt due to this game). Nowadays the market is clearly in a different mood. No one buys Treasuries on credit nowadays. It’s a rather over-hated investment these days. So if I would compare the current situation to any of the 3 last tightening cycles, I would rather see strong similarities to 2004, meaning I don’t see a large rise in Treasury yields, even if the Fed begins to tighten next year.
And stocks? They have been living a great life in the last 5½ years thanks to an ever-expanding Fed balance sheet.
It was not the bond market that gained from QE, but equities and all risky stuff like high yields, as I pointed out in post last year. See How QE Alters Bond Yields (Or Rather How It Does Not).
With the end of QE, the risky asset party will be over, while bond yields will be depressed for another decade, till inflation will eventually pick up after the retirement wave will have peaked around 2025.
Franz Lischka works as an asset manager in Vienna, Austria. You can visit his blog at http://franzlischka.blogspot.com
