Conducting state-of-the-art research is one of the ECB’s core tasks. Cutting-edge research and dialogue with academic economists have become all the more important as the number and complexity of issues relevant for the ECB have grown in the post-crisis world. With these themes as a backdrop, the research department of the ECB (DG-Research) brought together world-class economists and hosted the first ECB Annual Research Conference on 27-28 September 2016.1 In his opening remarks, ECB President Mario Draghi called attention to how the first ECB Annual Research Conference was deliberately more ambitious in scope than previous research workshops held at the bank.2 The purpose of the conference was to promote the discussion of topics at the frontier of research in monetary and financial economics, banking and macroeconomics, with a particular relevance for monetary and macroprudential policy. “Good research is the foundation of good policy,” he said, explaining that the aim of the conference is to bring together academia and central banks working at the forefront of economics.
The conference featured a selection of theoretical and empirical papers on a range of topics in macroeconomics and finance. Authors and discussants were from top economics departments and the ECB. Nobel laurate Eric Maskin of Harvard University gave the inaugural keynote lecture named in honour of EU founding father Jean Monnet. The event concluded with a policy panel on “Monetary policy and financial stability in a low interest rate environment” chaired by ECB Vice-President Vitor Constâncio that also included panellists Stockholm School of Economics Professor Lars Svensson, Federal Reserve Bank of New York Associate Director of Research and Statistics Tobias Adrian, Bank of England Monetary Policy Committee External Member Kristin Forbes, and National Bank of Belgium Vice-Governor Matthias Dewatripont.
Jean Monnet lecture: Strengthening European Monetary Union
In the Jean Monnet lecture, Professor Maskin discussed the question “Should fiscal policy be set by politicians?”3 He argued that economic and monetary union has been especially successful in promoting higher incomes and growth, and praised the ECB’s role during and after the financial crisis. Still, he argued EMU has a serious flaw – it has no centralised fiscal policy. Fiscal policy, like monetary policy, is a complex issue with delayed effects which are difficult to disentangle for ordinary citizens, Prof Maskin contended. Voters are likely to be better informed about value-based issues than those of a technical nature, he added. Elected officials wishing to gain or maintain power are in turn likely to ‘pander’ to less well-informed voters on fiscal policy issues, instead of only striving to do what is best. The fact that in an EU context, national leaders are only truly accountable to their own electorates and not to all EU citizens makes the consequences of ‘pandering’ even worse than in other democracies. To remedy this, Prof Maskin proposed that an apolitical, centralised body (with an institutional set-up perhaps modelled on the ECB) could set the overarching parameters, though not the redistributive details, of fiscal policy in the Eurozone. “A radical proposal, yes,” Prof. Maskin admitted, “but I think we can all agree that the Eurozone needs pretty substantial reform… we will all be better off if the Eurozone continues to survive and flourish. But this is not going to happen automatically,” he concluded.
Individual papers: The best of theory and empirics
Among the empirical papers featured at the conference, Joseph Vavra (Chicago Booth) presented work on “Regional heterogeneity and monetary policy”, documenting how regional heterogeneity in access to finance can influence the impact of monetary policy stimulus both on aggregate activity and on regional inequality. The paper finds that the 2008 announcement of large-scale purchases of mortgage-backed securities (QE1) in the US reduced mortgage rates led to buoyant refinancing activity, increased aggregate spending, and mitigated the downturn. At the same time, regional dispersion increased, driven by the lack of refinancing activity in the regions with the highest loan-to-value ratios and the largest house price and employment declines, where the policy impact was also smallest. However, the large house price decline and the high regional correlation between house price growth and income were specific to the Great Recession. The 2001 US recession, in contrast, was not accompanied by a house price decline, and the regional variation in house price growth and its correlation with income were also small. A theoretical model of collateralised mortgage lending that can reproduce some of the key features of the observed response to the 2008 credit easing confirms that, under conditions similar to those before the 2001 recession, the effectiveness of policy easing remains high and it reduces regional inequality. The paper concludes that constant monitoring of regional heterogeneity in collateral values and income is essential to correctly assess the efficiency of policy easing as well as its impact on regional inequality in the US.
Ulrike Malmendier (Berkeley Haas) presented another empirical paper, this one on “The making of hawks and doves: Inflation experiences and voting on the FOMC”. Building on previous work by the authors on the effects of personal experiences on household inflation expectations, the paper explores whether the lifetime inflation experiences of the members of the US Federal Open Market Committee (FOMC) have a measurable impact on their voting behaviour. As an alternative to rational expectations, the authors assume that two board members with different average lifetime inflation experience forecast different future inflation rates. Furthermore, recent inflation experience influences the expectation of younger board members more than their more experienced counterparts. Consistent with their theory, they find that variation in personal inflation experiences explains part of the variation in FOMC members’ inflation expectations reported to the US Congress. And when FOMC members decide about the interest rate target they collectively give approximately one third as much weight to their own inflation experience than they give to the staff’s inflation forecasts.
Amir Sufi (Chicago Booth) presented the third empirical paper entitled ‘Household debt and business cycles worldwide’. The paper documents a novel empirical regularity that is robust across a number of specifications, samples, and time periods: changes in household debt are strongly and negatively correlated with future changes in GDP growth over the medium run in a large sample of countries. Strikingly, the results are mainly driven by countries with an inflexible exchange rate regime. The magnitude of the estimated effect is large, too – a one-standard-deviation increase in the household debt to GDP ratio over a three-year period in a given country is associated with a 2.1 percentage point decline in GDP in the next three years. The authors find little support for a credit demand-driven explanation, such as anticipated increases in productivity. However, they find strong support for a supply-driven explanation whereby easier access to housing finance leads to an increase in housing debt. They also show that there are spillovers from one country to another, making the global effect of changes in housing debt larger than the sum of the direct effects for individual countries.
Three theoretical papers analysed the causes and consequences of low interest rates, and the role of monetary and fiscal policy at the lower bound. Pierre-Olivier Gourinchas (UC Berkeley) presented a paper on “Global imbalances and currency wars at the ZLB”, showing that very low interest rates and large capital flows can arise from an international imbalance in the demand and supply of safe assets across countries. These imbalances are malign when monetary policy is constrained at the lower bound as countries running high current account surpluses end up ‘exporting recession’, in particular to countries issuing reserve currencies. While exchange rate devaluations are beggar-thy-neighbour (‘zero sum’), safe public debt issuance and permanent increases in the money supply (‘helicopter drops’) are beneficial for all countries, as are increases in government spending.
In the presentation of his paper on “Monetary-fiscal interactions and the Euro area’s malaise”, Bartosz Maćkowiak (ECB) showed that a baseline New Keynesian model, augmented with multiple national fiscal authorities facing a ‘fiscal limit’, reproduces key features of the Eurozone macroeconomic data since 2008, as a product of purely self-fulfilling expectation shifts: the two recessions, the spike in government bond yields in 2011 to 2012, and the declining paths of inflation and the central bank’s policy rate. Furthermore, the model predicts that inflation need not return to the central bank’s objective of below, but close to, 2% in the medium run. Next, the authors argue that the introduction of non-defaultable ‘Eurobonds’ (convertible at maturity into currency at par, like ECB reserves) to finance centralised purchases of a fraction of national public debt, together with a fall in the present value of primary surpluses in countries hit by a recessionary disturbance such as the Global Crisis of 2008, would have tilted the balance in favour of more accommodative fiscal policy and led to better economic performance in the Eurozone. The authors point out that the similar favourable stabilisation outcomes can be achieved if the ECB directly purchases national public debt, so long as fiscal policy changes such that the primary surpluses flowing to the ECB fall in present value.
Iván Werning (MIT), presented his paper “Monetary policy, incomplete markets and bounded rationality”, and discussed how forward guidance on interest rates works in an environment characterised by financial imperfections hampering borrowing by households, and a particular form of deviations from rational expectations. In the benchmark model with complete markets and rational expectations, changes in future rates, no matter how far out in the future, are as powerful as changes in current rates. In the paper, bounded rationality is modelled by assuming that while households are perfectly aware of the announced path of interest rates by the central bank, they deviate from rational expectations in thinking through the effects of this path for endogenous variables such as aggregate demand. The paper argues that when both deviations are considered, incomplete markets and bounded rationality, the power of forward guidance is unambiguously reduced relative to the benchmark model.
Finally, the conference featured two theoretical works on corporate finance. Douglas Diamond (Chicago Booth) presented a paper entitled ‘Pledgeability, industry liquidity, and financial cycles’. The paper is motivated by the interaction between firms’ financing choices and business cycles, and asks two questions:
- Why are downturns following episodes of high valuations of firms so severe and prolonged?
- Why does lending and investment recover so slowly during downturns?
The authors build a model where debt contracts highlight the tension between an asset sale channel, whereby assets can be pledged today and later resold, and a control rights channel, whereby future cash flow can be pledged to enforce external claims. The model has more tractable empirical implications than other models of leverage cycles. The model predicts that there will be low pledgeability, high leverage, and low accounting quality during booms which are perceived as likely to continue. As a result, if a negative shock materialises after such booms, capital allocation will be inefficient, lending will be low, the recovery will be slow, and the recession will be severe.
A paper entitled ‘Bank resolution and the structure of global banks’ was presented by Patrick Bolton (Columbia University). The paper starts from the observation that the proposed framework for resolving global systemically important banks (G-SIBs) is based on two components: a single point of entry (SPOE), and total loss absorbing capacity (TLAC). A bank holding company will issue equity and long-term debt as loss-absorbing capital, and subsequently, losses in operating affiliates will be offset by TLAC write-downs on the holding company balance sheet. The authors analyse how this resolution framework works for a G-SIB operating in multiple countries. In particular, what are the regulatory implications of a situation where the bank holding company is in one country and the losses accumulate in an affiliate in another country? The main result is that the new resolution model is a big step forward, and even if it does not work as intended, lender-of-last-resort support will be much easier to justify following the write-down of TLAC. SPOE is the more efficient resolution mode, but regulators that are only concerned with the welfare of their own jurisdiction would limit its applicability. Therefore, SPOE requires cross-jurisdictional complementarities, and in the long run, stronger authority for supra-natural regulators. However, some open questions remain, such as what is the optimal TLAC, and who needs to trigger the resolution.
Panel discussion on monetary policy and financial stability
The closing panel addressed the topic of combining monetary policy and financial stability in a central bank – a topic of immediate interest to the ECB since it gained additional powers and responsibilities in the area of banking supervision. The panellists first discussed whether monetary policy should take financial stability considerations into account when the central bank is operating at the zero lower bound. There was agreement that expansionary monetary policy is needed as part of the current ‘optimal policy mix’ for the Eurozone. While QE helps banks by raising credit volumes and future growth, many banks are hurting from low interest rates in an environment characterised by slow nominal GDP growth (and high non-performing loans), post-crisis banking overcapacity, higher capital requirements and technological challenges (digitalisation). However, monetary policy cannot be the ‘only game in town’. It is necessary to also ‘protect’ monetary policy by avoiding financial instability. There was broad agreement that the first line of defence should be macroprudential policy, in particular in the face of asset price bubbles but especially to increase bank resilience. In this regard, one panellist argued that monetary policy can hardly achieve financial stability and thus should not have a financial stability goal. Using monetary policy to lean against the wind would only have a minor effect on financial stability but large costs in terms of macroeconomic stabilisation. A popular case for leaning against the wind was rebutted with the argument that even though the interest rate gets ‘in all of the cracks’, the empirical evidence indicates that a modest policy-rate increase will barely cover the bottom of those cracks. The logical conclusion was that in a weak economy close to its lower bound on interest rates, the costs of monetary policy leaning against the wind are even higher. Another panellist contended instead that monetary policy affects financial vulnerability by impinging on leverage, maturity transformation, and asset valuations both in the banking sector and in the shadow banking sector, by pointing to evidence consistent with a risk taking channel of monetary policy. In a model in which such a channel is introduced, even when optimal policy only cares about inflation and the output gap, it still acts as to reduce financial vulnerability. A second line of defence in the Eurozone was suggested by a panellist, namely to facilitate the ‘cleanup’ of the European banking system. A way to implement it would be to (temporarily) relax the ‘no-bailout constraint’ embodied in the Bank Recovery and Resolution Directive. The alternative approach – bail-in – could be risky in terms of financial instability, which would be extremely costly for public finances. Finally, the panel addressed the global financial stability implications of prolonged low interest rates in terms of capital flows. One panellist pointed to the lack of evidence that capital flows are currently excessive, or that asset price movements or GDP growth have become more correlated with the global cycle in EMEs. Therefore, countries seem on average to be better positioned to withstand swings in capital flows, in particular because of the widespread reduction in banking system leverage. However, concerns were raised on how low interest rates can amplify the impact of monetary policy through an exchange rate channel.
Overall, the conference marked an important effort to bring together an international group of policymakers and academics focused on topics at the frontier of policy discussions. The dialogue that occurred continues to advance the ECB’s goal of striving toward a more efficient and well-functioning economic and monetary union.
Authors’ note: The views expressed in this column are those of the authors and should not be attributed to the ECB or the Eurosystem.
Beraja, M, A Fuster, E Hurst and J Vavra (2016) “Regional heterogeneity and monetary policy”, University of Chicago.
Bolton, P and M Oehmke (2016) “Bank resolution and the structure of global banks”, mimeo, Columbia University.
Caballero, R, E Farhi and P-O Gourinchas (2016) “Global imbalances and currency wars at the ZLB”, mimeo, Harvard University.
Diamond, D, Y Hu and R Rajan (2016) “Pledgeability, industry liquidity, and financing cycles”, mimeo, Chicago Booth School of Business.
Farhi, E and I Werning (2016) “Monetary policy, bounded rationality and incomplete markets”, mimeo, Harvard University and MIT.
Jarocinski, M and B Maćkowiak (2016) “Monetary-fiscal interactions and the euro area's malaise”, mimeo, European Central Bank.
Malmendier, U and S Nagel (2016) “The making of hawks and doves: Inflation experiences and voting on the FOMC”, mimeo, UC Berkeley.
Mian, A and A Sufi (2016) “Household debt and business cycles worldwide”, mimeo, Princeton University and University of Chicago.
 The program and papers can be downloaded here. All presentations, including slides of the panellists, can be found here.
 The opening remarks of the ECB President can be found here.
 A video of the presentation can be found here.