Frank Smets is Director General of the Directorate General Research of the European Central Bank.
He is a Research Fellow of the Centre for Economic Policy Research in London and a managing editor of the International Journal of Central Banking.
He has written and published extensively on monetary, macroeconomic, financial and international issues mostly related to central banking.
Before joining the European Central Bank in 1998, he was a research economist at the Bank for International Settlements in Basel, Switzerland. He holds a PhD in Economics from Yale University.
Questions by TSE students
• Questions about research at a policy institution
How different is it for a researcher to be employed by an institution like the European Central Bank (ECB) instead of being a full time faculty member at an economic department of a university? Does the lack of teaching obligations give you more time to focus on your research activities?
I have never been a full-time faculty member myself so it is difficult to compare. One of the great things about working in the research department of a policy institution like the ECB is the direct contact with policy makers and the almost daily confrontation with new topical policy questions. When I started working for the Bank for International Settlements after graduating from Yale University in 1992, I found this the most rewarding aspect of becoming a researcher in a policy institution. At Yale, one of my frustrations was to come up with interesting, policy-relevant research questions. At the BIS the problem was the reverse: I did not know which policy question to address first.
As a researcher at the ECB one does not have teaching obligations (although many of my colleagues do teach on the side). However, one does have other non-research obligations such as writing and presenting policy notes, preparing a speech, participating in the projection process, being involved in the organization of seminars and conferences, etc. One of the challenges is that compared to teaching and particularly in crisis times these activities are a bit less predictable and may therefore interfere with one’s research commitments. As with teaching, the opportunity is to explore the synergies between one’s research activities and one’s policy contributions.
What if your research findings contradict some of the implemented policies supported by ECB? Do you feel restricted by the political nature of this institution?
An important objective of the research department is to provide input in the policy process. In management jargon, we want to be a “trusted source of research-based policy advice”. The ECB’s policies therefore partly build on research findings. In this context, it is also the research department’s role to “think out of the box” and to potentially challenge prevailing policies. It is therefore natural that different views are sometimes expressed. The failure of mainstream economists to foresee the crisis shows that more than ever it is important to avoid groupthink. Research plays an important role. At the same time, it is of course important that the publication of research findings do not interfere with the ECB’s communication and policy of a “single voice”.
That is why we have an ECB working paper series where most of our research is published under the responsibility of the individual authors. The editorial process focuses on ensuring sufficient quality and appropriate drafting, but is not there to stop the publication of policy sensitive research papers.
• Questions about DSGE modeling and policy in the context of the crisis
The Smets-Wouters model is part of the standard repertoire of the ECB, being the medium scale DSGE model that the ECB uses. The applications of DSGE (dynamic stochastic general equilibrium) style models have been widespread ranging from pure story-telling and computational illustration, to policy evaluations and forecasting.Wheredoyouseethefutureof DSGE modeling? Which particular trends and extensions would you emphasize? What applications might be more/less feasible in the future than they are now?
Medium-scale DSGE models such as the ECB’s New Area-Wide Model (NAWM), the New Multi-Country Model (NMCM) or the Christiano-Motto-Rostagno (CMR) model have over the past decade become standard tools at central banks and other policy institutions. In spite of all the criticism that these models have received following the financial crisis, I think that these tools are here to stay. Every model is only an imperfect and very partial description of reality and is usually designed for specific purposes and to analyze specific questions. It is therefore natural that models evolve as new questions come up. Most policy institutions use a suite of models not only to be able to address different questions, but also to robustify the policy implications.
I would emphasize two particular extensions. First, since the financial crisis a lot of research has focused on introducing financial intermediation in macroeconomic models. We have learned that the banking sector is not always a veil and we need to better understand how the financial system and the real economy interact. Under what circumstances does financial instability arise? How does it impair the monetary transmission mechanism? What does it mean for regulatory and non-standard monetary policies? Another direction for on-going research is a better modeling of how expectations are formed and how diverse views by economic actors develop and interact. Both of these developments are challenging because they introduce much more heterogeneity (they move away from the so-called representative agent paradigm) and non-linearity in the DSGE models. This makes solving the models more complicated and time- consuming, but as computer speed and solution algorithms advance new frontiers continue to be explored.
Yet these models and their use in policy making are also heavily criticized. What are the methodological shortcomings of this type of approach in your opinion? And what would you respond to the critics who think that such models when taken too literally may give a false sense of scientific sophistication?
As I mentioned before, any model is an abstraction of reality and, unfortunately, it takes a model to beat a model. The role of formal models in policy making is generally exaggerated: policy makers do not only use a suite of models, but also a lot of judgment, simple rules of thumb, intuition, etc. when making policy decisions. The standard New Keynesian linearized DSGE models with their emphasis on price and wage stickiness have and will continue to play a useful role, but they were not designed to address some of the new questions that came up in the financial crisis. Some of those questions such as the impact of financial fragility on the transmission process may be addressed in extensions of those models. For other questions, radically different models with different forms of expectation formation and heterogeneity may have to be developed. This makes macroeconomics so exciting in the current context.
• Questions about Macroeconomics and Central Banking in the wake of the financial crisis
There appears to be a sharp divide in the analysis of what caused the current European debt crisis, and how it should be resolved. For example, how do you view the German-French debate between austerity and growth? And what do you think of the contributions macroeconomists and central bankers have made to this debate?
Characterizing the current debate as one between austerity and growth is misleading and too simple in my view. The sources of the euro area sovereign debt crisis are multiple and not just the result of excessive sovereign debt accumulation or excessive consolidation. In Ireland and Spain the proximate source of the problems is the bust of the excessive credit and housing boom and the fall out in terms of a fragile banking sector, weak economic activity and falling government revenues. In Portugal and Italy the most important fundamental factor is the very low productivity growth over the past decade and a half. Only in Greece the underlying problem was excessive government debt accumulation. A common symptom in each of these countries has been a persistently large current account deficit and a large and increasing external debt of close to 100 percent of GDP. When the financial crisis led to a freezing of the interbank market and a sudden stop in capital flows from the core to those countries, this set in motion a number of “diabolic” negative spirals between the banking and the sovereign risks and between those risks and the real economy. Throwing more debt after debt cannot stop these negative spirals. Instead a combination of policies needed to be put in place. This includes i) fiscal consolidation in particular in those countries where the sustainability of government debt is put into questions; ii) structural reforms aiming at improving competitiveness to turn around the current account deficit, stop the accumulation of external debt and regain the trust of international investors; iii) a restructuring and recapitalization of the financial sector to allow it to provide credit to new projects; and iv) a credible backstop mechanism like the ESM and the ECB’s OMT to avoid that self-fulfilling fears of a break-up of EMU undermine government finances. At the same time, institutional reforms need to be undertaken in order to mend the design flaws in EMU and break the link between sovereign and banking risks. The most important reform in this respect is the establishment of a financial market union, comprising of unified banking supervision, a common bank resolution mechanism and a joint deposit insurance system.
The fiscal consolidation debate is not about austerity versus growth, but it is about how can governments ensure the sustainability of their debt and improve the efficiency of the government with the view of promoting growth. Fiscal consolidation needs to be done in a smart way and this may differ from country to country. Fiscal consolidation has long-run positive output effects. The question is how these medium-term positive effects can be brought to the present. The answer is by ensuring credibility of the program. In countries in which confidence is low this will require front- loading with possibly negative short- term output effects. Research on fiscal multipliers is important to calibrate those fiscal consolidation policies.
What lessons do you draw from the financial crisis? What are the important questions that you feel we (economists) need to work on to avoid similar events in the future?
For me one of the most important lessons is that we need to increase the buffers both in policy making and in financial markets.
In what direction do you think Macroeconomics will evolve? Will we see a shift of paradigms, much like we saw in the 70s? If so, what will the new macroeconomics look like?
Difficult question! There will be more focus on expectation formation and behavioral phenomena such as herding, groupthink and cognitive biases. This will require thinking about new dynamic equilibrium concepts. Macroeconomics will have to pay more attention to the financial system and its interaction with the real economy. This means that rather than one interest rate; macro models will have a multiple of interest rates. It also means that more emphasis will be put on stock and flow interactions and monetary and credit quantities. One of the stylized facts of financial crises across the globe is that credit booms almost always precede them. The interaction between monetary and fiscal policies will also come more and more to the forefront as central banks exit from crisis management.
Last but not least: What advice would you give graduate students who wish to start their own research careers in the area of business cycles and monetary policy?
It is an exciting time to be studying business cycles and monetary policy. The great recession and the euro area sovereign debt crisis will provide inspiration for many decades to come. I would advice students to look at European data and issues and not only to follow the American research agenda.