In this publication: Two broad metrics that capture the conditions of the US labour market have weakened in the past five months, albeit at a different pace. Despite the loss in momentum, Chair Yellen still sees a healthy labour market; other labour market indicators are still flashing green. We think that gains in nonfarm payrolls should recover, lifting the labour market conditions index, but will likely hover a bit below the average reported in the past year.160622-US-LMFCI-2.pdf (322 KB)
The Fed’s metrics to assess conditions in the labour market
Against the backdrop of the recent weak jobs market report, we think the slowdown in the jobs market warrants more attention. In a speech in Jackson Hole in 2014, Chair Yellen mentioned that she looked closely at two indicators that capture the labour market conditions. The first, the Fed’s Board Labour Market Conditions Index (LMCI), which includes 19 indicators, to a large extent included in the labour market report published by the Bureau of Labour Statistics. Among the indicators included are the payroll employment, labour force participation rate, workers classified as part time for economic reasons, hires and quits. The index does not include job openings, or data derived from business and consumer surveys that assess the pulse of economic activity. The second gauge Yellen said she followed closely, is a related index on the labour market published by the Federal Reserve Bank of Kansas. This metric includes 23 indicators and adds some of the employment surveys, such as ISM manufacturing employment that was omitted by the previously mentioned metric. In both indicators, a positive value indicates that it is above the long-term average.
Explaining the divergence between the two metrics
• On the one hand, the differences could be attributed to methodology. The inputs of
the Fed Board’s model are de-trended, while the other is only a measure of the first difference and therefore has weakened much less. Another essential feature of the Fed Board’s index is that is places greater weight on movements which are highly correlated with each other.
• Other differences are related to the composition of the broad indices. Lately, the indicators that have been included in the Federal Reserve Bank of Kansas’ index, but not in the other index, have been much more positive, including employment surveys. As a result of all this, the divergence between both indicators that make up the composite has been increasing in the past five months
Is the labour market conditions index effective in predicting a downturn?
Historically, the LMCI has fallen to negative territory on several occasions. This has not always led to a recession. A negative LMCI has signalled a recession in 40% of the time. However, since the 1980s all recessions in the US have been preceded by a negative LMCI. It’s fair to say that from all the components of the LMCI, nonfarm payrolls shows the highest correlation with the broad index, suggesting that nonfarm payrolls on its own is also a relatively good indicator for gauging the business cycle1. History suggests that six months before a recession, payrolls fall to an average of 160K, reporting at least two months of nonfarm payrolls below 100K. Most other components are lagging the LMCI. However, since 2013, the unemployment rate has also made large contributions to the improvement of the broad index.
Implications for the Fed
The implications of the downward trend of the LMCI in the past five months, however, have been almost neglected by Fed policymakers. Their recent communications even suggest that they still see an improvement in the labour market. Chair Yellen mentioned during the press conference after the June FOMC meeting, that the LMCI is a kind of experimental research, trying to play down its role. This is in sharp contrast to her view on this indicator when the LMCI was stronger. She added during the press conference that other indicators of the labour market, including jobless claims and job openings are still improving, although she has pointed to some downside risks that domestic demand could falter. The April’s FOMC minutes suggested that most participants thought that if the economy improved, this could support labour market conditions strengthening, with only a few mentioning that slower growth could have a negative effect on the labour market. This is understandable since the labour market lags economic activity.
Our base case for 2016 is that nonfarm payrolls will increase at a more moderate pace than in the past year, while unemployment should stabilise. We are more cautious regarding the labour market given that we see that the US economy is slowing and with it profit growth has weakened compared to the previous years. Going forward, the impact of a stronger USD and falling oil prices on profits will slowly fade, while we expect output growth to remain moderate in the second half of the year, basically driven by consumption. Consumption growth has been more upbeat lately. If consumption growth continues to improve, this should lift domestic demand. On balance, the chances of a rebound in profit growth rand strong hiring remain limited in the coming couple of quarters. This all suggests that job growth should recover and hover a bit below the average reported in the past year.
1. This can be observed by decomposing the change in LMCI into contributions from each indicator, holding the remaining indicators constant.