Analysis and Forecasting Department (France team)

The unemployment figures for the month of January 2016 published by France’s Pôle Emploi job centre show a fall of 27,900 in the number of job seekers who are not working (category A), which follows an increase recorded in the month of December (+15,800). While this fall might seem encouraging (a decline of this magnitude has not been seen since 2007), it must be qualified. First, recent changes in administrative practices made by Pôle Emploi [1] have resulted in an abnormal increase in exits from the jobless rolls due to failures to update (239,000, against a monthly average of 207,000 in 2015). Second, the high volatility of the monthly figures in recent months is a sign of a labour market in which job creation is insufficient to reduce unemployment on a sustainable basis.

It is true that the increase in the numbers exiting the job centre due to regaining work (+ 5.1% over three months) is a positive sign, suggesting that the expected recovery is underway. Nevertheless, even though a pickup in employment has occurred, it has not been strong enough to halt the steady rise in the number of long-term unemployed (+9.1% in one year). Thus, in a context of near-zero average growth since 2008 and a continuing deterioration in the labour market, the share of the unemployed registered for a year or more in categories A, B or C has increased since mid-2009 (by 31% approximately) and is now at a historical high, representing 45.4% of all jobseekers in categories A, B or C (Figure 1).

This increase is explained by the rise in unemployment among older workers (+ 8.9% yoy): the implementation of a series of pension reforms (2003, 2010), coupled with the elimination of job search waivers for seniors, has led to prolonging the working life and to a later retirement age. In a context of weak growth, the increase in the employment rate of older workers has been insufficient to absorb the growth in the working population in this age group, with a consequent rise in unemployment among those over age 50 (see La suppression de la Dispense de recherche d’emploi: quand les gouvernements augmentent volontairement le décompte des chômeurs ! [The elimination of job search waivers: when governments voluntarily increase the unemployment count – in French].

The relative improvement in the labour market expected in the coming months would stem from a slight improvement in growth and from the implementation of a training plan for the unemployed, announced by President François Hollande in late December 2015. However, it will take a long time for this improvement to affect the long-term unemployed. Indeed, the time taken for a fall in the numbers of Category A jobless to be transmitted to the long-term unemployed is relatively long (Figure 2). In the late 2000s, a period that saw a significant drop in jobless numbers, it took almost a year and a half for the fall in Category A jobless to result in a significant drop in the number of the long-term unemployed. The mechanisms for a pickup in jobs are clearly subject to considerable inertia.

[1] Because of this change in methodology, the unemployed have had one day less to complete their updates, leading in practice to a significant increase in the number of those struck off due to a failure to update (+1.5% in three months).

By Bruno Ducoudré and Pierre Madec

Considering the euro zone on the one hand and the United States and the United Kingdom on the other, changes in unemployment rates are a reflection of the divergences in growth highlighted in our last fiscal year forecast. While between 2008 and late 2010, trends in unemployment reflected the sharp deterioration in growth and did not differ much between the euro zone, the UK and the USA, differences began to emerge from 2011. In the United Kingdom and the United States, unemployment has been falling since 2011, whereas, after a brief respite, a second phase of rising rates took place in most euro zone countries (Table 1). It was only more recently that the unemployment rate has really begun to fall in Europe (late 2013 in Spain and early 2015 in France and Italy). Overall, for the period 2011-2015 the rate rose overall (+2.7 points) in Spain. In Italy, this deterioration in the labour market even worsened (+4.5 points in this period, against +2.2 points from early 2007 to late 2010). France, though to a lesser extent, was not spared.

An analysis of the unemployment rate does not however convey the full dynamics at work in the labour market (Tables 2 and 3), in particular in terms of underemployment. Thus during the crisis most European countries reduced the effective working time [1] to a greater or lesser degree, through policies on partial unemployment, the reduction of overtime, or the use of working-time accounts, but also through the expansion of part-time work (especially in Italy and Spain), including on an involuntary basis. Conversely, the favourable trend in the US labour market is partly due to a significant decline in the participation rate, which stood in the first quarter of 2015 at 62.8%, 3.3 points lower than eight years ago.

In order to measure the impact of these adjustments (working time and participation rate) on unemployment, it is possible, subject to a number of assumptions [2], to calculate the unemployment rate at constant employment and control for these adjustments. Except for the United States, where the participation rate has fallen sharply since 2007, all the countries studied experienced an increase in their labour force (employed + unemployed) that was greater than in the general population; in many countries this was due to pension reforms. Mechanically, in the absence of job creation, the impact of this demographic trend is to push up the unemployment rate in the countries concerned. For instance, if the participation rate had remained at its 2007 level, the unemployment rate would be lower by 1.6 points in France and 1.1 points in Italy (Table 4). Conversely, without the significant contraction in the US labour force, the unemployment rate would have been more than 3 points higher than what was seen in 2015. Also note that since the crisis Germany has experienced a significant drop in unemployment (-4.2 points) even though its participation rate grew by 2.2 points. Assuming an unchanged participation rate, Germany’s unemployment rate would be 3.1% (Figure 1).

In terms of working time, the lessons seem quite different. It thus appears that if working time had been maintained in all the countries at its pre-crisis level, the unemployment rate would have been more than 3 points higher in Germany and Italy and about 1 point higher in France and Spain, countries in which working time decreased sharply only from 2011. In the US and UK, the situation is very different: working time has changed only very little since the crisis. By controlling for working time, the unemployment rate thus changes along the lines observed in the two countries.

The tendency for working time to fall is a familiar story. Since the late 1990s, all the countries studied have greatly reduced their working hours. In Germany, between 1998 and 2008, the reduction was on average 0.6% per quarter. In France, the transition to the 35-hour week caused a similar reduction over the period. In Italy, the United Kingdom and the United States, the downward shifts in average working hours were respectively -0.3%, -0.4% and -0.3% per quarter. In total, between 1998 and 2008, working time fell by 6% in Germany and France, 4% in Italy, 3% in the United Kingdom and the United States and 2% in Spain, which was de facto the only country that during the crisis intensified the decline in working time that started in the late 1990s.

[1] Working time is understood here as the total number of hours worked by employees and the self-employed (i.e. total employment).

[2] It is assumed that, at constant use, a one-point increase in the participation rate leads to an increase in the unemployment rate. Employment and working time are not considered here in full-time equivalents. Finally, neither the “halo of unemployment” nor any possible “bending effects” are taken into account.

By Analysis and Forecasting Department (OFCE-DAP)

While the slowing increase in the number of job seekers registered with France’s Pôle Emploi unemployment agency in the first quarter of 2015 could be seen as the premise of the long-awaited downturn in the unemployment curve, the figures released today once again cast doubt on this prospect, at least in the short term. The registration of 26,200 additional people in category A at the agency in April brings the increase in job seekers back to a high rate, well above the average over the last two years (13,400 per month) and far from the virtual stability seen in the first quarter (+3,000 per month).

While the publication of strong figures for first-quarter GDP growth (+ 0.6%) reaffirmed the prospect of a recovery, the jobless numbers are disappointing. Don’t forget, however, that employment does not immediately respond to a pick-up in activity; it will take time to reap the benefits for the labour market of the good growth experienced at the year’s beginning, when the recovery has proven to be strong, pushing employers to recruit. For now, companies are still digesting the overstaffing inherited from the period of very low growth between 2011 and 2014. The fall in unemployment that can be foreseen with the recovery will not take place until the second half of 2015. But the acceleration of job centre registrations in April sends a contrary signal.

The situation seems to be generally worsening among all sections of job seekers: men, women, and all age categories. The number of unemployed under age 25 has been rising again the last months (9,500 people). But changes like these are often volatile, and should be treated with caution: they come in counterpoint to an equivalent fall in numbers during the first two months of 2015. Over a one-year period, the increase was only 11,900, and the interruption in the rise in youth unemployment since April 2013 signalled success for the jobs policy targeted at this group (see the figure). The announcement by the Minister of Labour of the creation of 100,000 additional subsidized jobs reflects the government’s perhaps belated determination to beef up this programme at a time when the economic outlook is improving.

By Sabine Le Bayon, Mathieu Plane, Christine Rifflart and Raul Sampognaro

Structural reforms aimed at developing a more flexible labour market are often attributed all the virtues of fighting against mass unemployment and limiting the segmentation of the labour market between “insiders” on stable contracts  and “outsiders” who are unemployed or on precarious contracts. When the economy is growing, these measures can facilitate job creation for the benefit of the outsiders, but the results are likely to be more uncertain in a context of mass unemployment and sluggish growth. Structural reforms can indeed reduce the labour market duality arising from regulatory measures but they cannot combat the duality of the labour market inherent in human capital, which is exacerbated during periods of mass unemployment: given the same qualifications it is experience that makes the difference, and given equal experience it is qualifications that make the difference. High unemployment therefore strengthens the phenomenon of “queuing” to access more stable jobs. Structural reforms aimed at streamlining the labour market will thus primarily affect employees who have less qualifications and experience without however enabling outsiders to gain access to more stable employment. This means that inequality between workers is likely to rise, with no positive impact on employment due to the sluggishness of the economy. Only macroeconomic management that takes on board the goal of returning to full employment could lead to successful structural reform.

As we show in a special study, “La dévaluation par les salaires dans la zone euro: un ajustement perdant-perdant” [Devaluation through wages in the euro zone: a lose-lose adjustment] (Revue de l’OFCE, no. 136, November 2014), labour market segmentation has increased during the crisis despite the implementation of structural reforms in the euro zone countries. Since 2008, the employment rate [1] of seniors and of the better qualified has fared better than for other population groups in the four largest countries in the euro zone (Figures 1 and 2).

The sharp decline in the youth employment rate since 2008 is general – including in Germany, where the labour market has remained dynamic – and contrasts with the increase in the employment rate of older workers (or the small decline in Spain). The difference between these two categories is between 12 percentage points in France and 21 points in Italy (15 points in Germany and 19 in Spain). The adjustment in the employment rate of the 25-54 age group lies in an intermediate position. The resistance of the employment rate of older workers to the crisis is probably due to a combination of two factors: the introduction of pension system reforms in recent years (lengthening contribution periods and / or raising the legal retirement age) and the relatively higher cost of dismissing senior citizens, who more often occupy higher positions in the job hierarchy. In a crisis, it is likely that this has led to a substitution effect with the employment of older workers coming at the expense of the young.

The adjustments in employment rates were also more striking for people without a high school diploma, with the exception of Italy, where the diploma does not seem to provide protection from unemployment or inactivity. In France, the adjustment in the employment rate clearly decreased with the type of degree. In Germany, the employment rate for those with less education has declined during the crisis while it has increased for the other categories. In Spain, the employment rate of university graduates has withstood the crisis better than the rate of other population groups. In addition to these developments in employment rates by educational category, wage income in Italy, Spain and France has fallen for the initial income deciles. This adjustment in the wage incomes of the lower deciles is probably due to a reduction in total working hours over the year (part-time work, shorter temporary contracts or longer periods of unemployment between contracts, reducing average compensation over the year). Thus, in the countries hit hardest by the crisis, the most vulnerable populations, with the least human capital, have found themselves more exposed to a deteriorating labour market, whether this has been felt through falling employment rates or a reduction in annual wage earnings.

In the context of a deteriorating labour market, by accepting a slight downgrade the most qualified unemployed workers would be the first to find jobs, chasing out those who might otherwise have gotten it, who would themselves do the same thing at a lower level. This could explain why, at the end of the queue, it is the least skilled who are, regardless of labour legislation, the victims of unemployment and precarious employment.

The existence of a “spontaneous” segmentation in the labour market and the phenomenon of “queuing” may thus limit the success of a strategy of structural reforms and wage devaluation. In such a case, a more flexible labour market combined with a reduction in social welfare could increase inequalities between groups in the workforce without increasing the creation of full-time equivalent jobs.

[1] This is the ratio of the employed to the working-age population.

By Christine Rifflart

Despite the further decline in the US unemployment rate in December, data from the Bureau of Labor Statistics released last week confirms paradoxically that the American labour market is in poor health. The US unemployment rate fell by 0.3 percentage point from November (-1.2 points from December 2012) to end the year at 6.7%. The rate has fallen 3.3 percentage points from a record high in October 2009, and is coming closer and closer to the non-accelerating inflation rate of unemployment (NAIRU), which since 2010 has been set by the OECD at 6.1%. However, these results do not at all reflect a rebound in employment, but instead mask a further deterioration in the economic situation.

While the unemployment rate is the standard indicator for summarizing how tight a labour market is, this can also be considered using two other indicators, i.e. the employment rate and the labour force participation rate – in the US case, these give a different view of the state of the labour market (see chart).

After falling nearly 5 percentage points in 2008 and 2009, the employment rate has been constant for 4 years, at the level of the early 1980s (58.6%, following a peak of 63.4% at end 2006). Since then, the decline in the unemployment rate has reflected the decline in the participation rate, a trend that is confirmed by the figures for December. Over the period 2010-2013, the participation rate lost a little more than 2 percentage points, to wind up at end December at its lowest level since 1978 (62.8%, following a peak of 66.4% at end 2006).

This poor performance is due to insufficient job creation, which has a threefold impact. Despite positive GDP growth – which contrasts with the recession in the euro zone – demand is far from sufficient to reassure business and revitalize the labour market. After four years of recovery, at end 2013 employment has still not returned to its pre-crisis level. Net creation of salaried jobs in the private sector has not even been sufficient to absorb the demographic increase in the working age population. As a result, the employment rate is not improving from where it bottomed out.

Moreover, the difficulty in finding employment is encouraging the exit or delaying the entry or return of people who are old enough to participate in the labour market. This effect, familiar to economists, is called effet de flexion (“bending effect”) in French: young people are encouraged to study longer, women stay at home after raising their children, and unemployed people become discouraged and stop looking for work. Despite the resumption of economic growth and job creation, this effect continued to be felt in full in 2013. While the reduction in the participation rate slowed in 2011 and 2012 – the growth of the labour force was once more positive but remained lower than that of the working-age population – it accelerated in 2013 with the decline in the labour force. During the second half of 2013, 885,000 people were in effect diverted away from the labour market, due in particular to the more difficult economic and social conditions.

Companies seem reluctant to rehire in the particularly difficult economic context. The fiscal shock in early 2013 depressed activity: GDP growth fell from 2.8% in 2012 to an expected level of about 1.8% in 2013. There will be additional fiscal adjustments in 2014. Beyond drastic cuts (related to sequestration [1]) in state spending, some exceptional measures that have been in force since 2008-2009 for the poorest households and the long-term unemployed (3.9 million out of the 10.4 million unemployed) are coming to an end and have not been renewed. According to estimates by the Centre on Budget and Policy Priorities (CBPP), 1.3 million unemployed who have exhausted their entitlement to basic benefits (26 weeks) and who have enjoyed an exceptional extension will find themselves without support as of 1 January 2014 due to the non- renewal of the measure, and nearly 5 million unemployed will be affected by the end of the year.

There is a risk of growing numbers of people falling into poverty in this situation. According to the Census Bureau, since 2010 the poverty rate has been about 15%. However, again according to the CBPP, unemployment benefits would have prevented 1.7 million people from falling below the poverty line. The greater difficulties facing the long-term unemployed and the withdrawal of part of the population from the labour market are the direct result of a morose labour market, which is not indicative of a continuous decline in the unemployment rate.

[1] See America’s fiscal headache written 9 December 2013.

By Eric Heyer

This text summarizes the OFCE’s October 2012 forecasts for the French economy.

The French economy is expected to see average annual growth of 0.1% in 2012 and 0.0% in 2013. This performance is particularly poor and far from the path that an economy recovering from a crisis would normally experience.

Four years after the onset of the crisis, the French economy has real potential for a rebound: this should lead to spontaneous average growth of about 3.0% per year in 2012 and 2013, making up some of the output gap built up since the start of the crisis. But this spontaneous recovery is being hampered, mainly by the establishment of budgetary savings plans in France and throughout Europe. The fiscal consolidation strategy imposed by the European Commission is likely to slice nearly 6 percentage points off GDP in France during 2012 and 2013.

By setting a pace that is far from its potential, the expected growth will increase the output gap accumulated since 2008 and will lead to a further deterioration on the labour market. The unemployment rate will rise steadily and hit 11% by late 2013.

Moreover, the reduction of the budget deficit expected by the Government due to the implementation of its consolidation strategy — the target for the general government deficit is 3% of GDP in 2013 — will be partially undermined by the shortfall in tax revenue due to weak growth. The general government deficit will come to 3.5% in 2013.

Under these conditions, should the government do whatever it can to fulfil its commitment to a 3% deficit in 2013?

In a context of financial uncertainty, being the only State not to keep its promise of fiscal consolidation is a risk, i.e. of being punished immediately by an increase in the financial terms on the repayment of its debt. This risk is real, but limited. The current situation is that of a “liquidity trap” and abundant savings. The result is a “flight to quality” phenomenon on the part of investors seeking safe investments. But among these are both German and French government bonds. Under these conditions, reducing the government deficit by 1 GDP point instead of 1.5 point would have very little impact on French bond rates.

However, maintaining a target of a 3% deficit in 2013 could have a dramatic impact on economic activity and employment in France. We simulated a scenario in which the French government maintains its budgetary commitment regardless of the costs and the economic situation. If this were to occur, it would require the adoption of a new programme of budget cuts in the coming months in the amount of 22 billion euros.

This strategy would cut economic activity in the country by 1.2% in 2013. It would lead to a further increase in the unemployment rate, which would reach 11.7% at year end, nearly 12%. As for employment, this obstinacy would intensify job losses, costing nearly 200,000 jobs in total.

A darker scenario is also possible: according to our forecasts, and taking into account the draft budget bills known and approved, no major European country would meet its deficit reduction commitments in 2013. By underestimating the difficulty of reaching inaccessible targets, there is a high risk of seeing the euro zone countries locked into a spiral where the nervousness of the financial markets would become the engine driving ever greater austerity. To illustrate this risk, we simulated a scenario in which the major euro zone countries (Germany, France, Italy and Spain) implement new austerity measures to meet their deficit targets in 2013. Adopting such a strategy would result in a strong negative shock to economic activity in these countries. For the French economy, it would lead to additional austerity that either at the national level or coming from its euro zone partner countries would cause a severe recession in 2013. French GDP would fall by more than 4.0%, resulting in a further increase in the unemployment rate, which would approach 14%.