5 Nov 2009.
The employment situation is key to the economy, both because of a direct impact on citizens, and because the overall health of the economy depends crucially on whether the single most important resource, namely labor, is being used effectively. It is generally agreed that the two key statistics, both from the Bureau of Labor Statistics (BLS), are (1) non-farm payrolls (NFP), a proxy for the total number of jobs, obtained by surveying employers, and (2) the unemployment rate, obtained by surveying individuals.
Next in importance is some measure of how much those employed are working. Some of the top measure are aggregate hours worked, overtime hours, and the count of those working Part Time for Economic Reasons (PTER; i.e. part-time even though they would prefer full time). While these all measure different things, they tend to move together, and it suffices, for a macro understanding of the jobs market, to watch one. We choose PTER.
PTER tends to be an early indicator of changes; so do two others that are worth watching: (1) change in the number of temporary jobs (again from the BLS), which tends to lead NFP by several months, and (2) initial unemployment claims, from the Department of Labor's Employment and Training Administration, which give a weekly read on current layoffs. Together these five indicators give a good sense of the overall employment situation. (Tutorial)
Part time for economic reasons background (29 Jun 2009) The count of those working “part time for economic reasons” is of interest (1) as one measure of unused capacity in the workforce and (2) as a leading indicator. The BLS define “part-time for economic reasons” (PTER) as “at work fewer than 35 hours [per week] for … reasons such as slack work, poor business conditions, or only being able to find a part-time job.” The basic idea is part-time but not by personal choice. The survey questions used to classify workers changed in 1994. The intent was to clarify rather than change the concept, but there was a large one-time change in the count. The YOY changes (but not the raw values) are at least roughly comparable pre- and post-change. Employers often cut back on hours before eliminating jobs altogether; on average the PTER YOY peak occurs a month before the end of the recession, the unemployment rate YOY peak occurs at the end of the recession, and the unemployment rate itself peaks 6 months after the end of the recession. Since the series started in 1956, previous YOY recession highs were in the range 27-64%. Multiple peaks have often been seen in previous recessions.
US unemployment claims background (20 May 2010) Initial unemployment claims are reported weekly and represent “real people standing in actual unemployment insurance lines … not based on samples with model guesstimates folded in” (Kasriel) and hence are a useful immediate indicator of emerging trends. The law on eligibility has changed substantially over time, and several other factors have brought about major shifts in uptake; thus the main emphasis should be on short-term trends. Claims are volatile; usually the YOY change or the four-week moving average is used; the YOY change has the advantage of at least roughly normalizing for changes in the rules and for population. Since coverage is short, the number of continuing claims, or 'insured unemployment', runs nearly parallel to initial claims; only one series need be followed. Initial claims tend to bottom several months before a recession starts and to peak at about the end of a recession. The YOY change in initial claims tends to lead the YOY change in non-farm payrolls by about 5 months, and to lead ISM manufacturing new orders by about three months. Various ranges of initial claims have been proposed as neutral for non-farm payrolls. None of these has strong support.
US non-farm payrolls background (24 Jun 2009) The non-farm payrolls (NFP) data series, from the BLS, is the best proxy available for total employment in the country, and is thus one of the most basic indicators of the overall employment situation. NFP are based on a monthly sampling, interpreted and reported in a matter of days, against a background of more comprehensive, but less timely, quarterly survey results. The monthly results necessarily involve some extrapolation (including the much-commented “net birth-death adjustment” for new businesses not yet reporting). While these extrapolations are reasonable, the MTM changes are quite noisy, and should not be given too much weight. The YOY change, with a neutral value of 1%, shows more clearly the trend. The YOY change is strongly correlated with, and lags, recession.
US temporary jobs (2 Jul 2010) The number of jobs in “temporary help services” has been increasing for several months – a positive sign for the job market. The YOY change in temporary help services, which typically leads changes in non-farm payrolls by about 4 months, has been positive for five months.
(Since the classification “temporary help services”, within the establishment survey, counts only those at temp agencies, while the Census probably hired workers directly, the impact of the decennial census on the temporary help services category was probably minimal.)
The current series from the Bureau of Labor Statistics began in 1990.
US unemployment background (11 Jul 2009) Most reported unemployment statistics are from the Bureau of Labor Statistics in the Department of Labor. All the BLS unemployment data are from the Current Population Survey, a monthly survey of a probability sample of approximately 60,000 households, conducted by the Bureau of the Census for the BLS. Data are collected by personal and telephone interviews. Response is voluntary, but only about 4% refuse. Roughly speaking, the BLS divides the population into the “Employed” (E), the “Unemployed” (U; must have looked for a job in the last four weeks), the “Marginally attached” (M; looking for a job but not in the last 4 weeks), those not looking for a job at all (N), and the institutionalized. The unemployment rate usually reported is U-3, including just the Unemployed (U / E + U). There is also a broader measure of unemployment, U-6, which includes the Unemployed, those working part-time for economic reasons (i.e. wanting full-time ; PTER), and the Marginally attached; these are all the people who are looking for full-time work but are unable to find it (U + PTER + M / E + U + M). There is one other statistic of interest, the “Not employed”. This includes those who are not looking for work (NLFW; so U + M + NLFW / E + U + M + NLFW). The survey questions and the definitions have changed over time. Probably the official unemployment rate has been pretty stable over time but, strictly speaking, current data is only comparable to data back to Jan 1994.
20 Jan 2007. Economist p33.
“In America around 20m jobs, or about one in seven, are lost involuntarily every year. Only a small fraction of those, some 2m-3m a year or 2% of all jobs, are permanent 'displacements', where workers have little or no prospect of returning to their old industry. The displacement rates for Europe are broadly similar.”