Employment growth is an important and underestimated macro factor of financial market trends. Since the expansion of jobs relative to the workforce is indicative of changes in slack or tightness in an economy it serves as a predictor of monetary policy and cost pressure. High employment growth is therefore a natural headwind for equity markets. Similarly, the expansion of jobs in one country relative to another is indicative of relative monetary tightening and economic performance. High relative employment growth is therefore a tailwind for the local currency. These propositions are strongly supported by empirical evidence. Employment growth-based trading signals would have added significant value to directional equity and FX trading strategies since 2000.
The below post is based on proprietary research of Macrosynergy.
This post ties in with this site’s summary of information efficiency and macro trends.
Employment growth and markets in a nutshell
Macroeconomic theory suggests that employment growth is related to future production costs, inflation, and monetary policy. A rapid rise in employment relative to workforce growth indicates excess demand for labour. This would normally lead to diminishing slack or increasing tightness of the labour market. Conversely slow employment growth or job losses would indicate diminishing tightness or increasing slack.
Tightening labour market conditions plausibly herald an upside shift in wage and labour cost trends, dampening profit margins for most companies. More importantly, from an overall market perspective, reduced labour market slack changes key parameters for the central bank: all other things equal the economy would need less monetary policy support while medium-term risks for the inflation outlook would tilt to the upside. In the U.S. employment is even an official goal of monetary policy.
If markets were fully information efficient, information on employment growth would only affect prices at the time of release. However, such efficiency is unlikely. The U.S. labour market report, typically published on the first Friday of each month for the previous month, is probably the single most important regular data release for global financial markets. This reflects that it is very timely and comprehensive with respect to the state of the economy. However, market commentary on the labour market report typically focuses on surprises versus expectations and supposed implications for the business cycle and monetary policy in the near term. There is less systematic attention to medium-term trends in employment relative to workforce growth and on a global scale.
Quantamental data of the market’s information state on employment
From a trading strategy perspective, the key question is if concurrently available information on employment growth helps predict future market trends and has historically served well as a signal for positioning in key macro markets. Here we investigate the use of employment trends for trading equity and FX markets in the developed world, as standard theory has clear propositions for the directional impact of these trends (see below).
In order to replicate the market’s information state on jobs trends, we use daily data of the J.P. Morgan Quantamental System (JPMaQS) for employment growth, as per cent over a year ago in three-month moving averages. View a description of the data set here. JPMaQS indicators use for each point in time the actual or approximate time series that the market observed at that time. This means it considers release dates and revisions and does so consistently across a range of currency areas.
In order to judge whether employment growth is high or low by a country’s standards, we subtract estimates of workforce growth. We call these differences “excess employment growth”. Positive excess employment growth conceptually indicates a tightening labour market, and negative excess employment growth (shortfall) indicates a slackening labour market. Adjustment for workforce growth or long-term employment is important as demographics can vary considerably across countries, even within the developed world. Thus since 2000 employment growth in the Antipodeans has averaged nearly 2%, while it was near zero for Japan.
The below panel shows the timelines for daily information states on simple and excess employment growth for ten developed countries or currency areas since 2000. The time series display pronounced cyclical effects, disproportionate variations around the time of the COVID pandemic and only limited short-term volatility.
The impact of jobs growth on equity markets
High employment growth should, on its own, indicate a shift of monetary policy towards higher policy rates, a shift in labour market conditions towards stronger wage trends, and a shift in the inflation outlook towards upside risks. All of these shifts would be unfavourable for equity markets. Hence, if financial markets are not fully information efficient, we would expect that concurrent information on positive excess employment growth forecasts negative broad equity market return trends while information on a shortfall in employment growth predicts positive equity return trends.
Empirical evidence since 2000 for developed markets strongly supports the proposition of a negative relation between excess employment growth and equity market trends. The target returns are the main local currency-denominated equity index future returns of eight developed market currency areas with sizeable equity markets:
- AUD: Standard and Poor’s / Australian Stock Exchange 200
- CAD: Standard and Poor’s / Toronto Stock Exchange 60 Index
- CHF: Swiss Market (SMI)
- EUR: EURO STOXX 50
- GBP: FTSE 100
- JPY: Nikkei 225 Stock Average
- SEK: OMX Stockholm 30 (OMXS30)
- USD: Standard and Poor’s 500 Composite
On a monthly look ahead basis there has been a significant negative almost 10% linear correlation between the employment factor and the main local currency-denominated equity index future.
This negative correlation has been stable across forward horizon frequencies (weekly, monthly and quarterly) and the observed decades. Also, the correlation has been negative for 7 of 8 developed markets. Japan has been the only exception. Japan’s special experience possibly reflects that the country struggled with deflation during most of the sample period. And with policy rates trapped near the zero bound, a shift towards tighter labour markets could have led to higher inflation expectations, lower real interest rates, and hence easier financial conditions.
Accuracy and balanced accuracy of excess employment growth as predictor of subsequent monthly equity index returns have been roughly 51% across all recorded months and countries. This may not look all that high but the accuracy of this indicator is suppressed by its lack of long bias. The employment signal does not consider or seek to earn any risk premia. Indeed, on its own, the excess employment growth signal would have implied a slight short bias in equity exposure since 20000, while equity index returns have been mostly positive, in 60% of all months to exact. One lesson is that employment and risk premia signals may best be considered in conjunction.
Next we calculate naïve PnLs based on the following assumptions:
- Positions are taken based on z-scores of employment growth shortfalls, winsorized at 2 standard deviations to avoid concentrated risk-taking in a single market.
- Positions are rebalanced monthly with a one-day slippage for trading.
- The long-term volatility of the PnL is set to 10%.
Note that this PnL is called “naïve” because it does not consider transaction costs. They might be very low for this strategy but ultimately depend on the size of assets under management. Also, the portfolio is not subject to proper risk limits or volatility targeting. The purpose of displaying a naïve PnL is to shed light on the pattern of value generation and its relation with the market portfolio.
The chart below suggests that the positive PnL generation of the employment signal (blue line) is very seasonal, focusing on times of drastic job cuts. This is a natural characteristic of a trading signal that stipulates no exposure or short positions in equity in normal or good times.
The features of the employment-based naïve PnL suggest that jobs growth is indeed a promising contributor signal to a directional equity strategy:
- The 2000-2022 (October) Sharpe ratio of the employment-based daily PnL has been 0.6 and its Sortino ratio 0.9. This is respectable for a strategy that has not been optimized at all, trades infrequently and incurs low transaction costs (and is therefore applicable for large assets under management). Whilst value generation is concentrated on times of labour market slumps, even in long periods of economic expansion and equity short bias it does not display a negative drift.
- The employment-based PnL displays little correlation with the long-only market portfolio. As an equal contributor to the management of a long-biased or cash-equity portfolio it would have more than doubled the Sharpe ratio of the fund from 0.4 to 0.9-1.0 from 2000 to 2022.
The impact of jobs growth on FX returns
High employment growth supports tighter monetary policy and greater tolerance for currency strength on the part of the central bank. It is also often indicative of strong competitiveness of an economy. Hence, we should expect that, all other things equal, countries with relatively strong employment growth post positive FX forward returns. Hence, the signals are information states of relative employment growth of a country versus employment growth of the currency area that it is mostly trading against (USD or EUR).
Target returns are 1-month FX forward returns of eight “smaller” developed market currencies. According to dominant market convention and economic position 4 trade against the dollar (AUD, CAD, JPY and NZD), 3 against the euro (CHF, NOK, and SEK) and one against both (GBP). Positions are standardized to 10% expected annualized volatility to prevent the empirical analysis from being skewed towards the more volatile forwards.
The theoretical impact of excess employment growth on FX forward returns is strongly supported by empirical evidence. The correlation of the latest information state on relative excess employment growth on subsequent FX forward returns on a monthly basis has been near 10% and highly significant.
Also, the positive relation between relative excess employment growth and subsequent FX forward returns has been significant for both the 2000s and the 2010s and 2020s separately.
We calculate naïve PnLs under similar assumptions as for the equity strategy check:
- Positions are taken based on z-scores of relative employment growth shortfalls, winsorized at 2 standard deviations to avoid concentrated risk-taking in a single market.
- Positions are rebalanced monthly with a one-day slippage for trading.
- The long-term volatility of the PnL has been set to 10%.
The naïve PnL (blue line in the chart below) shows fairly consistent and significant value generation. The Sharpe ratio since 2000 has been 0.7 and the Sortino ratio over 1. Correlation with the S&P 500 has been 9% and none of the value generation is due simply to being long the smaller currencies versus their USD or EUR benchmarks.
