What is the relationship between age and recession

Since the Great Recession: weaker link between economy and employment

In recent years, the development of gross domestic product and employment did not fit the idea that the demand for labor is significantly influenced by the economy. This article specifies to what extent and why the labor market has decoupled from the economy and which determinants have instead gained in importance

Companies adjust the number of employees and / or working hours, labor intensity and capital stock to cyclical fluctuations in the demand for goods. The last three also vary the productivity of an employee. But the more stable the productivity, the more the economy of the goods market is transferred to the labor market. In 1977/1978 strong economic growth of 6.6% led to an increase in employment of 2.2%. And when real GDP shrank by 0.4% in the 2002/2003 recession, the number of employees also fell by 1.9%. These examples illustrate the usual pattern in German economic history: employment grew more slowly or shrank more than GDP. This was accompanied by a steady increase in productivity, partly as a result of technical progress, partly because companies in times of economic downturn decided to lay off their capital rather than - often more expensive - to adjust their capital stock.

The most prominent exception to this rule is the Great Recession. In 2009 real GDP contracted by 5.6%; however, the number of employees remained practically unchanged. The recession was absorbed by a drop in productivity per worker (and about half of it by working hours). Such “labor hoarding” had already existed to a lesser extent, e.g. at the beginning of the 1980s.

However, the Great Recession marks not only an exceptional but temporary departure from productivity-driven growth. It also marks the beginning of a fundamentally weaker productivity development if one disregards the recovery effects of 2010. In other words: Measured against the mixed economic development since then, the increase in employment has been relatively strong. The number of employees increased by 2.1% in 2012/2013 - almost as in the 1977/1978 example. This time, however, there was only an economic stimulus of a meager 1.4% economic growth. Does that mean that economic growth has become more employment-intensive? Or does that mean that the economy no longer has as much impact on the labor market as it once did, but is being replaced by other factors or at least substantially supplemented?

Less cyclical effects on employment

To distinguish between these two possibilities, we related employment growth to economic growth and additionally modeled a component that is independent of GDP. The influence of GDP on employment - the Verdoorn relation2 - is expressed by a coefficient that varies over time. The smaller this Verdoorn coefficient, the weaker the business cycle translates into employment: if GDP rises, only a few new jobs are created; if GDP shrinks, few people lose their jobs. This time-varying coefficient provides information about which of the hypotheses formulated above applies. If the employment intensity of economic growth had been higher, the coefficient would have risen in recent years; if employment and economic growth had been decoupled, it would have fallen.

The rise and fall of the coefficient can be attributed to different reasons. These can be of a structural nature, i.e. they can create lasting effects. However, they can also be of a cyclical nature and only cause temporary changes. The time-variable coefficient therefore consists of a structural and a cyclical component (see Figure 1). After rising sharply in the early 1970s, the structural component hovered around 0.4 percentage points by the mid-1990s - an additional percentage point of GDP growth (contraction) increased (decreased) employment growth by 0.4 percentage points. A slight downward trend set in in the second half of the 1990s, which intensified abruptly at the time of the Great Recession in 2008/2009. This is one reason for the German “employment miracle”: the sharp slump in GDP was passed on to the labor market much more slowly than previously. This mitigating effect was exacerbated by the negative cyclical component, and the overall coefficient reached its lowest point during the Great Recession. Interestingly, the impact of GDP on employment did not return to its previous level when the German economy had long since recovered. The Great Recession left a permanent effect on the relationship between GDP and employment; it permanently loosened this relationship.3 Compared to the situation 20 or 40 years ago, the strength with which GDP growth impacts employment growth has halved .

illustration 1
Time-varying parameter for the impact of GDP on employment and its components

Source: own calculations based on data from the Federal Statistical Office.

The cyclical component fluctuates between -0.2 and 0.1. This means that the effect that an additional percentage point of GDP growth has on employment can be reduced by up to 0.2 percentage points or increased by 0.1 percentage points as a result of temporary influences. As a rule, the effect is intensified immediately before a recession (see the rise of the light blue line at the beginning of the gray shaded bars). Accordingly, for cyclical reasons, employment was adjusted somewhat more to recessions than to upswings. This, too, has weakened significantly since the Great Recession.

Why the employment intensity of GDP is changing

Sectoral change plays a prominent role for the structural component of the employment intensity of GDP. Because companies react differently to the economy, the heterogeneity between companies, sectors, etc. influences the macroeconomic impact of the economy. The service sector, for example, has lower productivity and lower volatility over the business cycle than industry. The more important services have become over the decades, the weaker the aggregate economic effect has become. Furthermore, the Great Recession has left a permanent mark on the economic structure: The trade / hospitality / transport sector has not made up for the value-added shares that were lost at the time; this also lowers the Verdoorn coefficient. Furthermore, less pessimistic economic expectations lead to fewer layoffs in a downturn. The companies keep their employees and thus limit the effects of a recession if they expect a relatively short or mild economic downturn.

The cyclical component of the GDP influence on employment also depends on the sectoral economic structure, which changes along production chains over the short term over the business cycle. In addition, the tension in the labor market (number of vacancies per unemployed person) dampens the impact on employment in recessions. The hoarding of workers is used here as a short-term strategy in order to smooth out the workforce in a tense labor market and to avoid long and expensive recruiting processes later. In addition, for cyclical reasons, employment fluctuates more strongly with GDP when the potential labor force grows faster, so that companies can choose from a larger supply of workers. In this sense, a growing number of jobs should be seen as a counterbalance to a tense labor market: It allows companies more leeway in adjusting their workforce. The growth contribution of investments or the number of patent applications as a proxy for technical progress were not significant in the regression analyzes.

GDP-dependent and GDP-independent employment growth

On average, GDP-related employment growth is 0.2% per quarter. This value shifts up or down if either the coefficient (Verdoorn effect) or GDP growth (GDP effect) deviates from its average. In addition, employment growth varies independently of GDP (autonomous effect). The three effects are shown in Figure 2. The employment impact of the Great Recession remained limited because the strong negative GDP effect was offset by an exceptionally large positive Verdoorn effect and equally strong autonomous employment growth. The subsequent recovery was relatively low in employment due to the small Verdoorn coefficient. The autonomous effect, on the other hand, has supported employment almost consistently since 2007. It proves that the good labor market development after the Hartz reforms was not a purely economic phenomenon. On the contrary, in the two most recent recessions, the clear GDP-independent employment growth counteracted the GDP effect. Influences independent of GDP usually increased the cyclical impact on employment. But now they were able to compensate for the unfavorable economic situation.

Figure 2
Breakdown into GDP-dependent and GDP-independent employment growth

Source: own calculations based on data from the Federal Statistical Office.

What explains employment if not the economy?

The cyclical part of employment growth is also driven by sectoral shifts, particularly the construction and service sectors. In addition, it can be seen that working hours are a substitute for employment - if each employee works longer, employment grows at a slower rate. As the price of work, rising wages also dampen employment growth.

On the other hand, a tight labor market is causing companies to stop regardless of the economic situation. These workers may not be productive at first - but the company will participate directly in the next upswing, as the staff has already been hired and trained. The tension in the labor market increased significantly in the upswing after the Hartz reforms and reached a new record at the turn of the year 2011/2012 - in the middle of the European recession. It essentially explains the great importance of the autonomous component. The labor force potential was identified as a factor driving employment with some delay. Without the increasing participation of women and the elderly and without the recently extraordinarily high level of immigration, especially from Eastern and Southern Europe, employment could not grow as strongly at the moment, especially since many immigrants succeed in taking up employment directly.

Policy conclusions

Trends such as sectoral change not only change the structure, but also the interrelationships between the economy and the labor market. According to our results, profitable labor market development is no longer just dependent on a successful growth policy, it is an independent part of it. Demand-side stimuli can mitigate the recession, but no longer to the same extent as in previous decades. Further factors must be brought into focus. It is important, for example, to support flexible educational paths throughout their entire professional life in order to protect employees from their knowledge becoming outdated in the course of structural change. It was also shown that tense labor markets prompt companies to build up jobs or to hoard workers. In the medium term, this also poses a risk: the labor market could develop too little dynamism to ensure the necessary creation and disappearance of jobs that will keep Germany a highly productive and competitive business location. Companies will need further strategies to counter potential skills shortages. Fields of action for politics in this context are again education and further training, but also migration and strengthening the ability to innovate.

  • 1 For detailed explanations, technical details and references, see S. Klinger, E. Weber: On GDP-employment decoupling in Germany, IAB Discussion Paper, No. 21/2014, Nuremberg.
  • 2 Named after the economist P. J. Verdoorn, who first quantified the relationship between GDP and productivity growth in 1949.
  • 3 That does not mean that the coefficient cannot rise again in future years, but that it would require a new impetus.