Albert Edwards is back and spreading realism as usual. First, the SocGen strategist takes on the increasingly growing divergence between the NFP payroll and household surveys.
The 85,000 decline in December?s non-farm payrolls jolted (briefly) the markets back to reality. For it had almost been forgotten in the post-November payroll euphoria that we remain in the midst of a long-lasting balance sheet recession. Yet surprisingly weak though the December payrolls were, this disappointment pales into insignificance compared with the massive 589,000 decline in employment as measured by the Household Survey (the monthly Employment Report contains surveys of both Households and Company Establishments each month). Typically the employment measure preferred by the markets is the payroll data collected from the Establishment Survey, as it tends to be less volatile on a monthly basis (but the unemployment rate data is derived from the alternative Household Survey).
Although the household measure of jobs is typically more volatile on a month-on-month basis, it is notable how it has seen a far more marginal improvement on a trend basis when compared to the payroll series (see chart below), and just how bad December?s outturn was.
Which of these indicators should one be following?
Typically, at turning points, the Household Survey measure of employment growth tends to be a leading indicator of the payroll data. Payrolls often get revised much later to fall into line with the Household Survey measure. One key difference between the two surveys is the difficulty the Establishment Survey measure of employment has in picking up what is going on in the smaller company sector. These diverging trends appear to be particularly wide at present (see chart below).
Some thoughts on why the ongoing weakness in the US economy continues being in the now forgotten small and medium business sector:
If you look at the ISM you would imagine that all is now well with the US company sector: indeed the December measure of new orders, standing above 65, is particularly strong. But like the Establishment Survey of non-farm payrolls, there is a bias towards larger companies. The evidence from the National Federation of Independent Businesses (NFIB) is that the situation for the smaller company sector is far gloomier, and to focus on the payroll survey is to miss the pain that is evident throughout the wider corporate sector in the US.
Why the birth/death model, which assumed 59,000 were created in December, is skewing data, and why the payroll survey can be effectively cast away as the latest data series indicator in the government's increasingly misleading and muddy representation of economic status:
One source of overstatement in the payroll survey occurs because the Bureau of Labor Statistics (BLS) makes assumptions about the birth/death of companies and the creation of new jobs - link. In December, for example, the BLS assumed 59,000 jobs were created (seasonally unadjusted). It is generally recognized that this model tends to overstate job creation in a cyclical downturn. It is also likely that, with the smaller company sector currently in so much pain, the payroll estimate of employment growth is overestimating jobs growth and that the wider ranging Household Survey may be a better indicator of current trends.
Some more on why the ignored small company sector is where the bulk of the action likely is, and why it continues being largely ignored by the BLS:
For it is quite clear looking at NFIB Survey that small companies are still having a terrible time. Nominal sales are poor and companies remain ?credit crunched?. In every month last year small companies planned to increase prices, whereas they actually had to slash prices. No surprise then that plans to hire and expand are still so very dismal.
Those working in the markets often forget just how important the unlisted corporate sector is to the economy overall. US companies with less than 500 employees make up half of all private sector employment, for example. The continued problems for the US smaller company sector might explain why the Chicago Fed?s measure of National Activity remains insipid. This monthly survey aggregates 85 of the key nationwide indicators and is thought by many to be a good monthly reading of the heartbeat of the national economy. A three-month moving average of around -0.7 is consistent with recession while a reading around zero is consistent with trend growth.
The latest reading in November is consistent with the economy only just beginning to emerge from recession. Yet already some of the leading indicators we monitor closely have begun to top out (e.g. ECRI and Conference Board). Furthermore the wonderfully named Philly Fed?s Aruoba-Diebold-Scotti high frequency measure of business conditions index already seems to be on the slide.
Yet where Edwards shines is in his observations of the divergence between the run up in forward P/Es and still all time low expectations of long-term EPS growth.
This means the equity market is far more reliant on the expectations for strong 2010 earnings growth being fulfilled (a near term eps disappointment can be shrugged off if valuations are dependent on high long-term earnings expectations). The 27% year-ahead expectation for global non-financial eps growth is a near record and particularly challenging when margins are at such high levels. The tentative evidence that the cyclical upturn may already be stalling out would leave an expensive equity market ready to complete the Ice Age secular de-rating.
Lastly, in debunking yet another rose-colored glasses myth, Edwards takes on the misperceptions surrounding the "surging" corporate cash flow and the barrage of M&A deals "just around the corner."
I hear much talk abound that the US corporate sector is throwing off unprecedented surplus cash and that it is only a matter of time before this triggers a major recovery of investment and/or M&A. Indeed both the household and corporate sector?s financial position seems to have been transformed from their noughties low of 4% deficits, with both sectors now running healthy surpluses of 3%
The national income accounts (NIA) definition though should be seen besides the Fed?s Flow of Funds (FoF) calculation of the very same surplus. On a slightly different definition this shows a more moderate surplus of only 1% of GDP. And when one assumes a neutral contribution from inventories (which will almost certainly be the case in Q4), the surplus drops away to 0% (see chart below). This is indeed certainly much better than a few years back, but not though compelling evidence that an investment/M&A boom will be imminent this year.
At the end of the day, nothing matters: fundamentals have long been pushed away to the scrapheap of relevant data. As Rosie keeps harping: "good news is good news and bad news is good news." What else is there to say? Maybe a few things. Our analysis on the upcoming Minsky moment will be published shortly.