It would happen, wouldn't it. One day after I stick my neck out far enough to say I think the equity markets are looking overpriced, and ready to take a knock, we get a string of very positive looking data, and off we go again. Well that's good, it's chastening to get humbled from time to time, but then, the game isn't over yet, not by a long stretch.
First it was the turn of the Conference Board to come in with an increase in its Index of Leading Economic Indicators, then the Department of Labour followed suit with a drop in new signings to below the dreaded 400,000 mark (386,000 provisional). Then the Philadelphia Fed waded in with its monthly industrial guage of manufacturing in the U.S. Mid-Atlantic region which leapt to 22.1 August from 8.3 in July (don't however miss the detail that the continuing higher productivity has allowed firms to do more with fewer workers, and thus delay new starts. The Philly Fed's employment index slipped sharply, to -8.7 from 0.8. ). At the same time various voices at the Fed are busy trying to reassure everyone that interest rates are set to stay low as far as the eye can see.
So where's the problem. Everything looks set to run, doesn't it. Well before we start why not look at a couple of charts. Firstly try this looks good, doesn't it? Now try this . A bit different, huh. Both the charts are of the NASDAQ composite, the first is over three months, and the second over five years. The perspective is different don't you think? What looks in the first to be a nice solid upward trend has another feel to it in the longer perspective. Sure we may be on the way up, but equally we could be about to run out of steam just one more time. Who knows? (BTW just glancing at the chart, the rebounds in the spring and autumn of 2002 seem to have had a lot more force to them than this recent one, but it's only a thought). So let's apply the same principal to industrial production. Maybe output has come back to life in the last three months, but that doesn't stop capacity utilization being still at 20 year lows. Then, and as everyone by now must surely know, there's the little problem of employment. Given the strong productivity performance and the increasing working age population the US economy may need to grow at around 4% to start generating net new employment. So where are we really? Paul Krugman puts his finger on something important in a recent NYT piece (Twilight Zone Economics):
Since November 2001 — which the National Bureau of Economic Research, in a controversial decision, has declared the end of the recession — the U.S. economy has grown at an annual rate of about 2.6 percent. That may not sound so bad, but when it comes to jobs there has been no recovery at all. Nonfarm payrolls have fallen by, on average, 50,000 per month since the "recovery" began, accounting for 1 million of the 2.7 million jobs lost since March 2001. Meanwhile, employment is chasing a moving target because the working-age population continues to grow. Just to keep up with population growth, the U.S. needs to add about 110,000 jobs per month. When it falls short of that, jobs become steadily harder to find. At this point conditions in the labor market are probably the worst they have been for almost 20 years. (The measured unemployment rate isn't all that high, but that's largely because many people have given up looking for work.) ............
Is relief in sight? Over the last few weeks two numbers have led to a spate of optimistic pronouncements. One is the preliminary estimate of second-quarter growth, which came in at a 2.4 percent annual rate — about one point higher than expected. The other is the rate of new applications for unemployment insurance, which has fallen slightly below 400,000 per week................. Just to stabilize the labor market in its present dismal state would probably take growth of at least 3.5 percent; it would take much more than that to return the economy to anything resembling full employment. Meanwhile, about those unemployment claims: somehow that 400,000-per-week benchmark has acquired a lot more significance in people's minds than it deserves. For example, claims came in at 398,000 yesterday — and this was treated as good news because it was (barely) below the magic number. Well, here's some perspective: since November 2001 new claims have averaged 414,000 per week. A number a bit lower than that might mean stable or slightly rising payroll employment — but as we've just seen, that's not nearly good enough. For comparison, in 2000 — a year of good but not great employment growth — weekly claims averaged 305,000. ............The best guess is that growth in the second half of the year will be faster than in the first half, possibly high enough to create some jobs, but not high enough to make jobs easier to find. In other words, in terms of what matters most, the economy will continue to deteriorate.
Source: New York Times
LINK
Now this argument is interesting, since, among other things, it highlights the important difference in demographics between the US and most of the rest of the OECD. Due to record immigration across the 90's the US working age population continues to increase at a significant pace. So in fact the US has precisely the opposite problem to the other 'industrialised' economies, finding jobs for all those extra young people hitting the labour market. (Digressing a moment, I think it's time to get away from the 'many people have given up looking for work' cliche - in Japan, in the US, in Europe - I'm sure it's true, but the more I think about it, the more I think we need some better metrics. The 'official' unemployment rates are just about meaningless in terms of understanding anything - of course they're wonderful for political debate. What we need to follow is the working age population, participation rates - across cohorts - dependency ratios etc. What would be nice to know would be how the active labour force was changing as a proportion of the working age population, although even this is not too helpful until we get a better purchase on what this expression means these days, what with young people studying longer and the retirement ages about to go up. It would also be interesting to know each month how the relative participation rates across cohorts were changing, and within the 'dependent' population to have a breakdown between young and old on an ongoing basis. Bottom line: we need something like a new 'Boskin' to work out some metrics for labour market analysis, with something like a composite index as the end product, but as someone likes to say, 'oh well, never mind', at least the absence of such a thing gives us economists plenty of margin for guesswork and speculation).
All this is then compounded by the fact that the economy, when it is working is highly productive (parodying the late Rudi Dornbusch: the US manufacturing sector isn't working too regularly, but when it is, boy is it productive). So really, more than any other OECD economy the US is 'condemned to grow', and this is where we hit the snags. As I have been flagging, all that outsourcing is fine, but you need value generating activities to pay for it, you need to be able to export, and to export (as the Japanese have been finding out) you need growth somewhere else. You also need a currency which is competitively valued, but it is here that we discover just how finely balanced the whole thing is. The dollar drops, US manufacturing begins to see export opportunities, the economy starts to rev up, people see growth on the horizon, and back comes the money looking for action with the undesireable consequence that the dollar starts to climb, and the export opportunities start to disappear, and............
Not only this, the dollar rise is deflationary in its impact, so the fed starts to get nervous, so...........
Just how highly stung everything is can be seen from the treasury and mortgage interconnect. The fed genuflects one way, and down comes the treasury market sending up long rates and choking the recovery, the fed genuflects the other, and everyone gets busy buying treasuries waiting for the unconventional tools. Well you wanted to know why I think the game isn't over: now you have it.