The productivity challenge for Europe
Lectio magistralis by Mario Draghi, President of the ECB, marking the 100th anniversary of the Deusto Business School,
Madrid, 30 November 2016
The euro area economy continues to expand at a moderate, but steady, pace and labour markets are gradually improving, including notably here in Spain. This gradual upward trend is expected to continue, not least owing to our monetary policy measures. But productivity growth has remained very subdued. In 1995, productivity growth in the euro area was on a par with the rest of the world at about 2%. But now, at below 0.5%, it lags behind the growth rates of the United States, other advanced economies and emerging markets.
If it persists, this slowdown in productivity growth will matter greatly for our future prosperity, and will have direct consequences for the conduct of monetary and fiscal policy and the cohesion of the euro area.
Consider first the challenge caused by ageing populations, which many economies will face in the coming decades. The share of working-age persons in the total population is projected to decline across Europe, and this is particularly marked in some economies.
In the absence of change – if output per worker, structural unemployment, and labour participation remain at current levels – population ageing will result in a stark fall of output per capita. Using the OECD’s population projections, the decline in output per head by 2050 would be 14% in Germany, 16% in Italy and 22% in Spain. While this scenario is very unlikely to materialise, it shows that current rates of productivity growth are barely sufficient to offset the demographic drag on per capita income.
Structural reforms are therefore urgently needed to raise productivity growth and unlock unused labour potential and thereby avoid stagnation in per capita income.
But the benefits of structural reforms to the welfare of people in the euro area extend beyond increasing aggregate incomes and offsetting the demographic drag. They also empower other policies.
Stronger potential output growth aids monetary policy by increasing the equilibrium real interest rate. Higher equilibrium rates provide more room for monetary policy to operate without being constrained by the effective lower bound. This in turn reduces the likelihood that unconventional measures will be required to counteract future downturns.
And higher future growth helps monetary policy today. It encourages households to spend more and firms to invest, reducing the need for monetary policy to support current economic activity and bring inflation back towards 2%, and speeding up the return to more conventional monetary policy settings.
There are benefits for fiscal policy too. By lifting output and employment and lowering unemployment, reforms improve governments’ structural balances. Moreover, higher levels of potential output reduce the current overhang of public sector debt that is impinging on some countries’ ability to carry out stabilisation policies. The greater fiscal space also enables governments to redistribute the benefits of reforms across the whole population. Some reforms can have upfront negative distributional effects, which governments may want to offset.
Structural reforms also help fully unlock the benefits of the Single Market and thereby improve the cohesiveness of the monetary union. Increased convergence not only reinforces the integrity of the union. It also builds up the trust among Member States that is necessary to take further integration steps.
And finally, structural reforms make the euro area economy more resilient in the face of the disruptive influence of globalisation and digitalisation.
All these reasons highlight the vital importance of structural reforms for the welfare and well-functioning of our union. Let me now briefly consider the factors behind the euro area’s recent lacklustre productivity performance, before setting out some necessary steps for improvement.
Factors contributing to low productivity growth in the euro area
There are two underlying sources of labour productivity growth. The first is the amount of capital available to each worker. Raising investment is therefore a key part of tackling the productivity challenge in many advanced economies. The second relates to how efficiently resources are used, and includes such things as innovation and organisational improvements. This is total factor productivity.
It is in this area that the euro area performance has been particularly weak. There are many reasons put forward for this weakness, but in essence they relate to the role of innovation, technology adoption and efficient reallocation of resources.
Some argue that innovations now are on a smaller scale than in previous decades and centuries. The argument goes that the 19th century witnessed the diffusion of transformative inventions such as the railway and electricity; whereas nowadays our innovations are things such as an app to order a taxi, rather than phoning. To an extent this may be true, but the facts seem to say otherwise.
If falling total factor productivity growth were caused by slowing progress at the technological frontier, you would expect productivity growth of firms at that frontier to be correspondingly poor. The opposite appears true. From 2003 to 2013, productivity at global frontier manufacturing firms – i.e. the top 100 firms globally in each sector in terms of productivity – increased by a quarter. Productivity growth at global frontier service sector firms was even higher. And, importantly, the performance of euro area firms at the frontier is on a par with that of other advanced economies.
It is productivity growth among the non-frontier firms that is poor across the OECD countries, explaining the slowdown in aggregate productivity growth across advanced economies. Productivity in non-frontier manufacturing firms increased by just 10%. And, for the euro area, the picture for non-frontier firms appears even worse, with non-frontier productivity stagnating in manufacturing and even declining in services.
Overall, the picture appears to be not so much a lack of technological progress per se, but a lack of diffusion of progress from the frontier to other firms. In other words, the laggard firms aren’t absorbing the new technology and incorporating it into their production processes to increase their efficiency, possibly due a lack of competitive pressures.
But labour productivity growth not only depends on the efficiency gains of each firm in the economy. It also depends on the extent to which capital and labour are allocated to the most productive firms. Here, again, there are signs of underperformance in the euro area, particularly in the non-tradable sectors. Recent research suggests that this misallocation of capital and labour is trending upwards in many euro area countries.
In conclusion, therefore, the euro area’s poor productivity appears a result of poor diffusion of technology from the frontier to other firms, and from an inefficient allocation of capital and labour to relatively low productivity firms. Such factors are fortunately more receptive to structural reforms than innovation, which is harder to institute. There are two main areas that need to be addressed.
First, barriers to knowledge diffusion need to be brought down. This includes fostering a competitive business environment that encourages the adoption of the best management techniques and organisational structures. Here, there is a clear role for institutions such as the one whose centenary we are celebrating today to aid this diffusion of knowledge and best practice across firms.
Second, productive firms need to be able to grow, to attract both capital and skilled workers. This requires well-functioning capital, product and labour markets. It is crucial that the Single Market is completed, particularly in the services sector where misallocation is largest. Regulatory barriers that inhibit small firms from gaining headcount need to be relaxed – there is evidence that productive firms choose not to grow rather than cross the threshold and incur greater administrative expenses. Recent reforms in Italy and Spain have sought to address this problem.
Improving allocation also requires a well-functioning financial system that can direct funds to dynamic firms. Adequate institutional and judicial frameworks, such as bankruptcy laws and the ability of banks to enforce collateral arrangements, are also needed to ensure capital is not trapped in unproductive or failed firms. The steps taken here in Spain to reduce the legacy effects of the banking crisis will help invigorate the flow of new credit towards profitable and growing firms.
Bolstering the contribution from labour supply
The solution to the demographic challenge lies not just in making current workers more productive, but also increasing the working share of the population. Structural unemployment is high in the euro area, and has increased steadily for decades with each successive economic cycle. Labour force participation rates across much of the euro area similarly lag those of the top performers in the OECD by a large margin. Reducing the structural unemployment rate and increasing participation rates to the average levels of the top three economies in the euro area is in many cases sufficient to offset the demographic drag, even assuming output per worker remains the same.
In the short run, bringing additional workers into the workforce may lower average productivity. But lowering structural unemployment helps protect future labour productivity. Workers who remain unemployed for long periods suffer permanent losses throughout their life, in terms of lower income, productivity and employability. This is particularly the case for young unemployed people, who risk losing important experience in their formative years. The current high rates of youth unemployment are likely compromising the future productivity of this generation – and it’s precisely this group which will have to support a larger number of dependants than previous generations did.
The labour market reforms in Spain in mid-2012 are an example of a structural reform that has been successful in unblocking the labour market. The unemployment rate peaked at 26.5% in 2013 but is now at 19%. While still very high, the decline of nearly 8 percentage points in three years against a backdrop of major deleveraging is remarkable.
The European Commission estimates the reforms had a cumulative impact of 0.5% on GDP and of 0.8% on employment between 2013 and 2015. The OECD estimates that the reforms contributed to moderated unit labour cost growth, increased hiring by 8% and reduced job separation by 24%. Analysis by the IMF reached similar conclusions. Other examples of successful labour market reforms in the euro area have similarly resulted in an employment-intensive recovery in the reforming countries.
Labour markets need to improve to deliver a skilled, flexible workforce. Reducing skill mismatches lowers structural unemployment and enables productive firms to grow. It also importantly addresses one of the main factors explaining the increase in income inequality. The key here is education, which includes life-long learning and retraining, potentially bolstered by schemes to facilitate re-allocation of workers across sectors. Breaking down the barriers that create duality in the labour market will enable the unemployed youth of today to become the high-skilled older workers of the future.
The slowing momentum of reform
So, structural reforms to boost productivity and participation and reduce structural unemployment are vital. And that is why the Governing Council continues to call for such reforms in the euro area. Yet reform momentum appears to be slowing. Why then, given the clear benefits and urgent need for reforms, are governments reluctant to carry them out?
One factor is the distributional impact of reforms, particularly in the near term, which can make them unpopular. But what is often overlooked in the public debate about structural reforms is the distributional impact of inaction. The high rates of structural unemployment, duality in the labour market and people being discouraged from entering the labour market have also a human as well as an economic cost. Reducing unemployment is socially positive from a distributional point of view.
But it is clearly important to design the reforms so that they are as inclusive as possible. By counteracting any negative distributional effects of reforms, governments can build broad-based support for further welfare-enhancing reforms. Of course, not all redistribution resulting from reforms is socially negative. Removing inefficiently high economic rents in protected sectors, notably non-tradable ones, can have marked positive benefits on social welfare. Entrenched special interests can have powerful lobbying operations, though, that governments need to resist.
As these barriers to implementing structural reforms are perennial, what then has driven the implementation of reforms in the past? The literature on the subject is relatively new, but highlights a number of drivers, in particular economic crises and high unemployment. For the EU, the Single Market directives are an important driver of product market reforms.
Correct sequencing of reforms can also help – product market reforms can facilitate the subsequent implementation of labour market reforms. By eliminating barriers to entry, pro-competition product market reforms can eliminate inefficiencies, in turn promoting higher economic activity and employment, and thereby offsetting possible short-term costs of later labour market reforms.
But what about monetary policy? As it turns out, the literature says little about how monetary policy drives the reform process. One view is that an accommodative monetary policy stance may increase the room for manoeuvre of governments, allowing measures to offset some of the short-term costs of reforms. But there are certainly some commentators who at present believe that the ECB’s unconventional measures are stifling incentives for reform.
To answer this question, preliminary internal research at the ECB investigates the drivers of structural reform in 40 OECD countries over the past three decades. This research supports previous findings in the literature on the drivers of reform. On monetary policy, it finds that lower interest rates, if at all, tend to support rather than hinder the implementation of reforms. We also see this confirmed by anecdotal evidence in some larger euro area countries, which recently implemented some important labour market reforms.
It seems, then, that we should put more weight on the former view. By acting in accordance with our mandate, namely to support activity to bring inflation back in line with our medium-term objective, the ECB is in fact also providing space for euro area governments to carry out the necessary structural reforms. It is a window of opportunity they should seize.
Let me conclude.
The euro area faces a significant demographic challenge in the coming decades. Reversing the decline in productivity growth and improving labour market outcomes are both required to meet this challenge. Without concerted effort on structural reforms, per capita income growth in the euro area is likely to stagnate, and may even decline.
The welfare benefits of structural reforms extend beyond just protecting income per capita. The distributional impacts of lower unemployment and higher participation are positive. Increasing potential output growth provides both monetary and fiscal policy more space to operate in. By breaking down barriers, reforms allow the full benefits of the Single Market to be realised, increasing economic cohesion and strengthening the monetary union.
The ECB’s current monetary policy settings are designed to support activity to bring inflation back to our objective of close to, but below, 2% over the medium term. In doing so, we are lowering the risk that the current low rates of growth become entrenched, but we alone cannot eliminate that risk. Monetary policy is providing support and space for governments to carry out necessary structural reforms. It is up to euro area governments to act, individually at national level as well as jointly at European level.
Draghi, M. (2015), “Structural reforms, inflation and monetary policy”, speech at the ECB Forum on Central Banking, Sintra, Portugal, 22 May 2015.
E.g. Gordon, R. (2016), The Rise and Fall of American Growth: The US Standard of Living since the Civil War, Princeton University Press.
OECD (2015), The Future of Productivity, OECD, Paris, France.
Source: CompNet (2016), “European firms after the crisis: New insights from the 5th vintage of the CompNet firm-level-based database”.
Gamberoni, E., Giordano, C. and P. Lopez-Garcia (2016), “Capital and labour (mis)allocation in the euro area: some stylized facts and determinants”, ECB Working Paper No 1981.
See, e.g., Anderton, B. et al. (2015), “Comparisons and contrasts of the impact of the crisis on euro area labour markets”, ECB Occasional Paper Series No 159.
See, e.g., Ball, L. (2009), “Hysteresis in Unemployment: Old and New Evidence”, NBER Working Paper No 14818, March 2009.
European Commission (2016), “The Economic Impact of Selected Structural Reform Measures in Italy, France, Spain and Portugal”, Institutional Paper 023, April 2016.
OECD (2014), “The 2012 Labour Market Reform in Spain: A Preliminary Assessment”, the Organisation for Economic Cooperation and Development, Paris, France.
IMF (2015), “Spain: selected issues”, IMF Country Report No 15/233.
European Commission (2016), op. cit.
See, for example, Høj, J., Galasso, V., Nicoletti, G., and T-T. Dang (2006), “The Political Economy of Structural Reform: empirical evidence from OECD countries”, OECD Economics Department Working Papers No 501 and Agnello, L., Castro, V., Jalles, J. and R Sousa (2015), “What determines the likelihood of structural reforms?”, European Journal of Political Economy, 37: 129-145.
See, for example, Thompson, W. and R. Price (2009), The Political Economy of Reform: Lessons from pensions, product markets and labour markets in ten OECD countries, OECD, Paris, France.
Gordon, R. (1996), “Macroeconomic Policy in the Presence of Structural Maladjustment”, NBER Working Paper Series, No 5739