Selected Takeaways from the ECB’s Sintra Forum on “Inflation and Unemployment in Europe”

Vítor Constâncio, Philipp Hartmann, Oreste Tristani 28 October 2015

a

A

Shortly after its establishment in the historical setting of Sintra, near Lisbon, the ECB’s Forum on Central Banking has become a treasured retreat for policy makers, academics and market economists to contemplate a topic of common concern in a deep and encompassing way, looking beyond day-to-day pressures. In this post we summarise four of the main themes that were hotly debated in Sintra in 2015: the relevance of hysteresis for European growth and employment; the implications of the shape of the Phillips curve for policy; the need for innovation and productivity for economic prosperity; as well as the conjunctural implications of structural reforms.

1. Hysteresis redux

The Sintra papers by Olivier Blanchard et al. and Jordi Galí brought renewed attention to the hysteresis hypothesis. Blanchard, Cerutti and Summers (2015) provide empirical evidence that post-WWII recessions in many advanced countries are followed by lower trend growth than before the recession. This finding and its implications for the fact that high cyclical unemployment increases the equilibrium unemployment rate have already been emphasised in the literature of the 1980s (first in Blanchard and Summers 1986). The authors therefore confirm them also for the 2000s. Chart 1 illustrates the case of Italy, which is representative for a number of European countries. The authors conclude that, in the presence of hysteresis, deviations of output from its optimal level are much longer-lasting and thus more costly than usually assumed.

Figure 1. Evolution of log real GDP and extrapolated trends for Italy

Note: Vertical bars denote recession dates. Dashed lines denote the estimated trends, plus one-standard deviation bands associated with uncertainty about the value of the estimated trend coefficient.
Source: Reproduced from Blanchard et al. (2015), online appendix, Figure 3A.

Galí (2015) explains evidence of a stochastic trend in euro area unemployment since the 1970s (in contrast to mean-reverting U.S. unemployment) by embedding the insider-outsider phenomenon in labour markets in a New Keynesian macroeconomic model. Wages are set such that insiders remain employed without consideration of the fate of the unemployed (outsiders). A key implication of this theory is that there is no “natural” level to which the unemployment rate tends to gravitate over long periods of time. It can also account for the stability of wage inflation over the last two decades.

Many conference participants concurred that hysteresis effects play a role in Europe and should be taken into account by public policy. Some Sintra participants argued that such effects justify aggressive monetary policy actions to avoid recessionary episodes, which would have long or permanent effects under the hysteresis hypothesis (e.g. Ball 2015 or Summers 2015). It was also mentioned that hysteresis implies that monetary policy could produce effects on medium-term growth, not just dampen cyclical fluctuations.

Willem Buiter (2015) pointed out that the euro area suffers both from deficient aggregate demand and from fundamental supply side problems. Whilst the insider-outsider problem does not necessarily call for demand policies, he argued that to close the output gap the euro area needs an effective combination of monetary and fiscal policy, including the use of “helicopter money”. Demand policies could also somewhat help in reducing insider-outsider problems by increasing the number of insiders. But addressing this specific problem at source would require labour market policies, notably constraints on collective bargaining arrangements. Bob Gordon (2015) questioned whether the evidence for a stochastic trend in euro area unemployment is robust after the 1980s. As of the 1990s unemployment seems mean-reverting, albeit to a higher mean, and the attention should rather turn to the average level difference with the U.S.

2. Slope of the Philips curve and its implications

In line with the title of the 2015 Sintra Forum, a lot of attention was also devoted to the Phillips curve, in particular the strength of the relationship between unemployment and inflation that it implies. The evidence presented by Olivier Blanchard et al. (2015), Larry Ball (2015), Jordi Galí (2015) and Bob Gordon (2015) suggests that the slope of the Phillips curve flattened between the mid-1970s and the early 1990s. Since then it remained roughly stable. A very flat Phillips curve can obviously rationalise the “missing disinflation” during the Great recession, i.e. the relatively stable inflation rates recorded in many industrial countries in spite of the large increase in unemployment rates. The structural causes of the flattening are, however, less well understood. Dennis Snower (2015) reviewed a wide range of theoretical results indicating that the curve can be highly nonlinear and thus have different slopes in different segments. Moreover, Mario Draghi (2015) and Marco Buti for the euro area and Haruhiko Kuroda for Japan reported about estimations that the Phillips curve may have steepened again in recent years. Draghi (2015) related this to the increased responsiveness of inflation to cyclical conditions in countries that had reformed their product and labour markets.

Most of the discussion focused on the policy implications of an extremely flat Phillips curve. A number of speakers drew the inference that, in order to steer inflation towards levels consistent with price stability, monetary policy should react more aggressively to real economic conditions. Others countered that monetary policy can continue affecting prices via other channels, including the exchange rate and its impact on expectations.

3. Innovation and productivity

Chris Pissarides (2015) shifted the attention towards long-term growth. He started from the observation that the only sustained way to improve living standards is through continual innovation-enhancing productivity. He and various other speakers documented how the euro area, in particular Southern European countries, had fallen behind the U.S. – as the “world centre of innovation” – in terms of productivity growth or stopped catching up with it (Draghi 2015, Fernald 2015). Chart 2, which is taken from Fernald’s discussion, is showing this for a measure of labour productivity. Catherine Mann (2015) showed evidence for OECD countries that a lack of diffusion of productive innovations from frontier firms to non-frontier firms is part of the problem.

Figure 2. GDP per hour relative to the United States (US=100)

Note: Country comparison in U.S. dollars using purchasing-power-parity exchange rates.
Source: Conference Board. Reproduced from Fernald (2015), Chart 1.

John Fernald (2015) argued that the widely debated productivity slowdown in the U.S. after the dot-com bubble does not look like a sign of secular stagnation, but rather like a return to trend after the large increases during the 1990s. He contrasted this experience with the case of Europe, where productivity declined in recent years without increasing significantly in the 1990s. An important element of the U.S. productivity acceleration in the 1990s and early 2000s was, first, production in Information and Communication Technology (ICT) and, subsequently, ICT use in other industries. As for example Chart 3 – taken from Mario Draghi’s (2015) introductory Sintra speech – shows, still today the euro area is far behind the U.S. in ICT adoption.

Figure 3. ICT adoption (index: 1-7 (best); euro area is unweighted average)

Source: World Economic Forum Networked Readiness Index 2015. Reproduced from Draghi (2015), Chart 8.

Pissarides (2015) went on to highlight the benefits of the U.S. model for creating a good environment for innovators, characterised by ample university budgets and high industry research and development (R&D) spending. He then compared the U.S. model to the situation in Europe, which suffers from lower university budgets and cuts in infrastructure spending, notably in countries under fiscal stress. As a result, in terms of domestic expenditure on R&D, the EU28 average is not only lagging behind both the United States and Japan, but also losing ground to emerging countries like China, India or Korea (Fernald 2015).

Productivity growth tends to lead to net employment losses in the innovative sectors, but then new jobs tend to be created in other sectors, like services. For this job reallocation mechanism to work, however, a high degree of flexibility in product and labour markets is needed. Low growth of the services sector in Europe could be related to the high regulation of this sector in many European countries, including in Germany.

4. Structural reforms and their conjunctural implications

The discussion of structural problems in Europe sparked a lively debate about the nature, timing and sequencing of structural reforms relative to cyclical policy measures. Mario Draghi (2015) set the tone in the opening speech in which he highlighted the value of structural reforms in increasing the flexibility of the economy in response to shocks, reducing hysteresis effects and enhancing its long-run growth potential. They could unleash the “untapped potential” of euro area countries and help in making the ongoing cyclical recovery a stronger, structural recovery. The benefits of flexibility are particularly high in a monetary union, where large structural divergences across countries can become “explosive” and endanger the integrity of the union. The time for accelerating structural reforms now in the euro area is advantageous, because monetary policy meaningfully buttresses demand and fiscal policy is broadly neutral. Many Sintra participants supported the view that labour and product market reforms are needed for reducing European unemployment and for better preparing European countries for the structural change that productivity-enhancing innovation requires (including the contributions by Fernald 2015, Mann 2015 or Pissarides 2015).

Tito Boeri and Juan-Francisco Jimeno (2015), however, took a more critical view of the ongoing European structural policies and adjustment programs. They argued that reducing firing costs and the generosity and coverage of unemployment insurance is exceedingly costly during recessions. Better would be rule-based countercyclical unemployment benefits. Gilles Saint-Paul (2015; also supported by Marco Buti), however, pointed out that for political economy reasons reforms are rarely done in good times and cuts in unemployment benefits would also be hard to enforce in such times. Moreover, even if they are implemented in bad times, they are still beneficial as long as their discounted benefits are positive.

Boeri and Jimeno (2015) also proposed a number of labour market reforms at the European level, including a European employment contract with individual accounts (e.g. for severance pay) transferable across countries, European unemployment insurance and cross-border transferability of pension rights. Christoph Schmidt (2015), Olivier Blanchard and other participants, however, were of the opinion that the large majority of reforms should be done at the national level. Insider-outsider problems and skill mismatches received particular attention. Chart 4 – taken from Catherine Mann’s (2015) panel intervention – highlights the problems that could be addressed by structural reforms and is suggestive of the type of reforms that would have promise in improving the matching of skills in labour markets.

Figure 4. Framework policies and the probability of skill mismatch in labour markets

Note: The blue dot is the average probability to have skill mismatch, evaluated at the median level of the policy and individual characteristics. The distance between the Minimum/Maximum of the relevant policy indicator and the median is the change in the probability of skill mismatch with the respective policy change.
Source: McGowan, M., and D. Andrews (2015), Labour market mismatch and labour productivity: evidence from PIAAC data, OECD Economics Department Working Paper, No 1209, Paris. Reproduced from Mann (2015), Chart 7.

In our judgement the 2015 Sintra discussions suggest that solving Europe’s problems with unemployment and low inflation requires a two-handed approach, involving both demand-side policies and structural reforms.

All papers, discussions and speeches can be found in the conference e-book (ECB 2015) and video recordings of all sessions on the Sintra Forum website.

References

Ball, L (2015), Comment on “Inflation and activity” by Olivier Blanchard, Eugenio Cerutti and Lawrence Summers, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 47-52.

Blanchard, O and L Summers (1986), Hysteresis and European unemployment, in S. Fischer (ed.), NBER Macroeconomics Annual, MIT Press, Cambridge (MA), pp. 15-77.

Blanchard O, E Cerutti and L Summers (2015), Inflation and activity – two explorations and their implications for monetary policy, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 25-46.

Boeri, T and J F Jimeno (2015), The unbearable divergence of unemployment in Europe, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 117-144.

Buiter, W (2015), Unemployment and inflation in the euro area: why has demand management failed so badly?, ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 87-98.

Draghi, M (2015), Structural reforms, inflation and monetary policy, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 12-24.

European Central Bank (2015), Inflation and Unemployment in Europe, Proceedings of the ECB Sintra Forum on Central Banking, Frankfurt am Main, October.

Fernald, J (2015), Comment on “Structural perspectives on European employment: the role of innovation and growth” by Christopher Pissarides, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 164-172.

Galí, J (2015), Hysteresis and the European unemployment problem revisited, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 53-79.

Gordon, R (2015), Comment on “Hysteresis and the European unemployment problem revisited” by Jordi Galí, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 80-86.

Mann, C (2015), Structural perspectives on European employment and growth in a global context, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 177-184.

Pissarides, C (2015), Structural perspectives on European employment: the role of innovation and growth, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 152-163.

Saint-Paul, G (2015), Comment on “The unbearable divergence of unemployment in Europe” by Tito Boeri and Juan Jimeno, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 145-151.

Schmidt, C (2015), Designing and communicating structural reforms in the euro area: the unequivocal responsibility of Member States’ governments, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 193-200.

Snower, D (2015), A fresh look at the inflation-unemployment trade-off, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 104-111.

Summers, L (2015), Current perspectives on inflation and unemployment in the euro area and advanced economies, in ECB, Inflation and Unemployment in Europe, Frankfurt am Main, October, pp. 112-116.

a

A

Vice-President, European Central Bank

Deputy Director General Research, ECB; and CEPR Research Fellow

Senior Adviser in the Directorate General Research, European Central Bank

Events