Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
By Franz Lischka, CFA
Last week I wrote a commentary with the provocative title US economy: below stall speed or rather already above potential? I argued that the retirement wave of the baby boomer generation, after its first couple of years, will lead to much slower economic growth in the coming 10 to 17 years. What I subsequently noticed, was that my thesis was interpreted to be more negative than I intended. Or at least was the feeling I had after businessinsider.com reprinted the commentary with what appeared to be a Great Depression-era picture of a row of men looking for work. I suddenly found myself in the doom-and-gloom camp (or at least I felt so), the "the US is gonna be another Japan" and "another Great Depression is coming" crowd. Ironically, my my message was partly just the opposite, namely that unemployment will be the bright spot in the future and a lack of labor will be the problem.
Further, from those who read the article and posted a comment on businessinsider.com I found a lot of disbelieve and occasional hostility, asking whether I think a more restrictive monetary policy would be good given the current unemployment levels. First, let me say that the trend that I am talking about, the baby boomer retirement wave, is in a very early stage. Assuming a retirement age of 65 only the first 2-year age cohort, those born in 1946 and 1947, have now retired. Given that birth rates only started to cool after 1958 and were above average until 1964, another 10 to 17 years of boomer retirements will follow!
What I'm trying to do is identify trends for the coming years, which in my opinion are not yet common knowledge and therefore not priced into the markets. And let me add that I am an investor, not an economist. I don't want to tell the Fed or the US government what they SHOULD do (nor do I think that they would take my advice anyway). Rather, I try to figure out what they WILL do. it was the Fed's statement that it intends to keep interest rates at zero as long as unemployment is above 6.5% that triggered my interest in US labor market trends in the first place.
Another reason why I felt misunderstood to be a doomsayer, is that I wanted to make the point that a full-blown recession is not necessarily eminent, as some people argue (a bit contradicting today's title above, I know. I will explain later). How often have you heard and read in the last couple of years that "every time this or that indicator has fallen to such a low level, the US economy was in a recession or just entering one." And then the recession never came. But these assertions keep popping up on a regular basis. That's simply because in the past every time the US grew at such a slow speed as it is now (below 2%), a recession followed. But as I argued in my previous commentary:
For a further discussion on the topic I want to post a chart taken from the invaluable St. Louis Fed web site that Anton Vrba, a regular contributor on dshort.com, sent me in response to my article. It's the US population of 25 to 54-year olds:
What is so fascinating, or shocking, about this chart is that it shows what happens when the baby boomers leave a certain demographic group. The 25-54 cohort currently comprises people born between 1959 and 1988, which means that the majority of baby boomers have now moved out of this group, starting in around 2000, with another five years for a complete exit. In effect, the strong growth trend began leveling off at around 2000 and for the last five years has even became negative. What is so important is that we're talking about the entire population, not just the labor force. The shrinking labor force participation rate of the last couple of years is often attributed to people giving up looking for work, which I would argue has actually become a less significant factor over the past two years. Assuming an average retirement age of 65, we can forecast that the US working age population will look somewhat similar on an 11-year fast-forward basis, no matter how motivated people will be in looking for work.
What I'm suggesting is that by the end of the decade, assuming there won't be a huge wave of immigration, the labor force might be shrinking not only on a relative basis compared to the population but also on an absolute basis.
That would mean that some common assumptions will be plain wrong, especially the level of job growth that would be needed to keep the unemployment rate steady. Currently it is estimated that the monthly US nonfarm payrolls would have to be somewhere between 150 and 200 thousand per month to achieve that goal. In reality, given current population trends, that number in the later decade may actually be negative!
I ran an analysis of the relationship between monthly nonfarm payrolls and the year-over-year change in the unemployment rate in which I adjusted payrolls by the size of the US population to bring them to the December 2012 level. It resulted in a steady-state unemployment rate with payrolls growing by about 180 thousand (the point at which the regression line crosses the x-axis). But the interesting point is where the last three years happened to be on the chart:
Now as you can see, it is not entirely uncommon for the unemployment rate (y-axis, inverted) to fall significantly with payroll growth just at break-even levels. It actually happens quite often, usually in the year after a recession, probably a result of people giving up looking for a job. So 2010 was not an exemption, rather the norm. What is unusual is that this has now happened three years in a row (by the way, 2013 is off to a very similar start). This is better illustrated in the next chart, where I did the same on a yearly basis, but with a rolling three-year average. As you can see, the last three years are an outlier, giving support to my thesis that the break-even rate for the change in payrolls to keep the unemployment rate stable is now already at a much lower level (in the chart more to the left). Given the above mentioned trends in the population, I expect that level to decrease further in the next 17 years (perhaps to negative levels), shifting the future regression line much further to the left.
What does that imply for the future? This is difficult to envision because there hasn't been anything similar in the past. Perhaps the best way to get a feeling for future possibilities not by a similar event, but by the exact opposite, namely the so-called "jobless recovery", a term that, if I remember correctly, was popularized in the 1992 election campaign. It was not that the there was no job growth at all during the tenure of George H.W. Bush from 1989 to 1992, but given the labor force growth in the 80s and 90s it was much too weak and so unemployment shot up during what was not all bad, but rather mediocre economic growth, resulting in a losing reelections campaign for the elder Bush in 92. (I once read a very interesting article about the strong link between the change in the unemployment rate during an election year and the likelihood of the election of the incumbent's party's candidate)
Now what we currently might see is something of the exact opposite. People are regularly complaining about mediocre growth, but at the same time the unemployment rate has fallen 2.5% in the last 3 1/2 years, not that bad by historical standards and nothing that can actually last forever. Now let me explain what I meant by the title of this article, for which I coined the term "full-employed recession". It is not a recession by normal standards that I would expect. Perhaps the term "full-employed slump" would be more accurate, but I felt the earlier term was easier to bring my point home. What I would envision for the next 10-17 years is very mediocre US growth (especially once the unemployment would have fallen to around 6%), so low that just by the usual fluctuation in growth, negative GDP growth could happen from time to time without the usual strong deterioration in many economic indicators, especially with unemployment somewhere around the natural rate. Whether that could be defined as a recession is good question. By the standards of the NBER, it is unlikely. Productivity growth would probably keep overall growth positive on average even when the labor force should fall, but the growth will be quite low nevertheless.
Let's have a look at the unemployment rate of the past years and what that would mean for the future, especially for Fed policy. People are no doubt focused on the 6.5% target from the Fed statements, but the next big question is when the Fed might end their Quantitative Easing policies. Whereas ZIRP (zero interest rate policy) has its famous 6.5% target, as long as inflation stays low (and no guarantee that rates will rise then, just that they will not before), the Fed has been shy so far to state a target for the end of QE. In the following charts I assumed a level of 7%. I took that from some earlier remarks from James Bullard, the President and CEO of the Federal Reserve Bank of St. Louis, that he made at the beginning of the year, which implied that 7% might be an appropriate level to end QE. Further Janet Yellen, who is seen as Ben Bernanke's most likely successor, said in early April that "there will likely be a substantial period after asset purchases conclude but before the FOMC starts removing accommodation by reducing asset holdings or raising the federal funds rate." So if 6.5% is the level at which the Fed starts thinking about a hike in rates, at least in my imagination 7% looks as "a substantial period" before that. But please regard the 7% line more as a gray area than as a strict line.
For a longer perspective:
Given current trends at present, the end of QE seems to be a likely possibility in early 2014. What will happen at that point is an interesting guess. After the end of QE1 in spring of 2010 and QE2 in summer of 2011, markets saw a lot of turmoil and economic indicators nosedived. But it is worth noting that unemployment never really picked up during those periods; it just kept stable, which is what I assume will happen again after the end of the current QE programs. So I wouldn't bet that the trend will continue at its current speed in 2014. But then it will all depend on what the Fed will actually do.
The Fed may start another round of QE, if markets get too nervous, what then could actually bring the unemployment down to 6.5%, where they would at least have to consider a rate hike, what would at least at current speed would be in autumn of next year. Interesting to see is that the unemployment rate has now already come down more than half its way to the level associated with full employment (the natural rate of unemployment estimated by the CBO), currently at 6%. The structural problems associated with the rise in long-term unemployment that I mentioned last time is clearly reflected in the rise of this estimated natural rate since the Great Recession as can be seen above. At current speed that would be reached in late 2015, a time that I had considered in my last comment from the projected decline in the continuing claims to historical lows.
So what to make of all that? Is it all gloom and doom? Well, that depends on your perspective. If you are a job-seeker in the US who has found the notion of a jobless recovery frustrating, then I might have some good news for you. Although it may not feel that way yet, good times may lie ahead. But if you are an investor, than maybe not so. For investors jobless recoveries were actually not bad at all. They meant a growing economy and a very accommodative Federal Reserve. Well, the coming decade might mean the opposite: slow growth and at the same time as the end to easy money. Not good, but certainly not the Great Depression stuff that some imaginative editor might have associated my earlier analysis.
Franz Lischka works as an asset manager in Vienna, Austria.