Sabine Lautenschläger: Monetary policy - end of history?
The views expressed in this speech are those of the speaker and not the view of the BIS.
In 1992, the political scientist Francis Fukuyama announced the end of history. The cold war had just ended, and in Fukuyama’s view, this marked “the endpoint of mankind’s ideological evolution”. Liberal democracy had prevailed; the final form of government had been reached. There was no need for history to continue.
Similar claims were made about monetary policy. The great moderation, for instance, was traced back to better monetary policy. With this in mind, some argued that monetary policy had reached a perfect and final stage. There was no need to evolve further; history had come to an end.
Or had it? Well, it seems that the unexpected always happens. And when it does, it tends to push the end of history a bit further away. This is true for politics, and it is true for monetary policy.
The unexpected happened in 2008: the global financial crisis struck and undid the great moderation. History took a sudden turn and this threw long-held beliefs overboard. Academics as well as policymakers had to adapt their thinking and their doing – monetary policy was no exception. Central banks around the world came up with new tools to keep the financial system and the economy afloat. They became key players; some observers even referred to them as the only game in town.
Now, ten years later, there is one thing that weighs on many people’s minds: When will unconventional monetary policy end? When will it return to normal? It seems that many look no further than the exit from our unconventional measures. But what will happen once we have reached the exit? Will we return to the end of history?
Well, I don’t think so. First, there are still many questions that need to be answered. And second, the unexpected will happen again at some point. History does not end. But before we discuss what will happen tomorrow, let’s take a look at today.
Approaching the exit – slowly but surely
For many months, we have been discussing the conditions for a gradual exit from our extraordinarily loose monetary policy. But some might now point to the latest economic data indicating that growth has lost some momentum and wonder if this is reason to be concerned, or even to change plans?
Well, first of all, we need to put the latest developments into perspective. Growth in the euro area exceeded potential growth for a couple of quarters. In 2017, the euro area economy grew by 2.5%, compared with potential growth estimates of 1.5% for the year.1 Against this backdrop, one could expect to see it to slow down. And that’s what happened. In the first quarter of 2018, year-on-year growth was 2.5%, compared with 2.7% in the fourth quarter of 2017. We saw industrial production, excluding construction, decline in the first two months of the year. And both the composite output Purchasing Managers’ Index and the European Commission’s Economic Sentiment Indicator eased in the first quarter. But both are still well above their long-term average.
So we are seeing that the pace of growth has become more moderate, but we are not seeing a turning point. We remain confident in the strength of the economy.
After all, the things that are currently holding back growth seem to be temporary. There was the early timing of the Easter break, there was a strong outbreak of flu in some parts of the euro area, there was cold weather and there were strikes in some countries. All this weighed on growth, but it won’t do so permanently.
We need to keep a close eye on all this, and we will. But for now, there is no need to rewrite the story. The economic expansion remains solid and broad-based. Financing conditions are good, the labour market is robust with a historically high increase in jobs, and income and profits are growing steadily. In short: the real economy is doing well.
By contrast, inflation so far does not seem to be recovering as convincingly. This has left many observers scratching their heads as to why the current level of low inflation does not match the current state of the real economy. It seems that inflation is responding less to the slack in the economy than would be expected. This disconnect between the real and nominal sides of the economy is the subject of intense debate. In very general terms, there might be two forces at play.
First, the Phillips curve might have changed. It might, for instance, have flattened, or it might have shifted downwards. Empirically, it is very hard to determine which – if either – of the two things has happened.
And that brings me to the second point, which is that we cannot be sure whether we are measuring slack correctly in the first place. The unemployment rate, for instance, is based on a narrow definition. Just think of people who work part-time. Officially, they are employed, but they could work more, of course. So, the amount of slack could be larger than we think. If that is the case, it’s no surprise that inflation has not kicked in.
That said, the economy is still expanding and, at some point, any remaining slack will be used up. As a consequence, pressure on wages will increase, and we are indeed already seeing wages edging up. Wage growth increased from 1.1% in mid-2016 to 1.8% in the fourth quarter of 2017. In Germany, wages even grew at 2.7% in the first quarter of 2018. This, in turn, will prompt inflation to pick up.
So I believe that all we need is a bit of patience. All the conditions for inflation to kick in are in place.
So, June might be the month to decide once and for all to gradually end net asset purchases by the end of this year.
But please do not misunderstand me here: I am just referring to the net purchases. There is still the stock of assets we have already bought. And we will still reinvest any proceeds from maturing bonds. Just to put it in perspective: each month we buy assets worth €30 billion. This compares with a stock of assets worth €2,400 billion.
What I want to say is this: even when we end net purchases, monetary policy will still be extraordinarily accommodative. And, at the same time, other tools will remain in place. The Eurosystem has, for instance, committed to granting banks unlimited access to liquidity against a wider range of collateral. With our targeted longer-term refinancing operations, we provided credit institutions with funding for up to four years. Finally, interest rates are still historically low and will remain so for some months to come. A first hike around the middle of 2019 is not entirely out of the ballpark.
So, we are slowly but surely moving towards the exit. This is the next big thing to happen. But it will not be the end of the story, let alone the end of history. Monetary policy will continue to evolve, new challenges will arise, and many questions still need to be answered.
I will discuss three of these questions now: First, does monetary policy need to refine its targets and update its toolbox? Second, how should it treat financial stability? And third, what is the relationship between monetary policy and the distribution of wealth and income?
Tools and targets
As we know, when the crisis hit in 2008, central banks around the world had to rethink their approach in more ways than one. And they are still doing so.
What will monetary policy look like in the years to come? Should we put some of the tools we used in the crisis near the top of our toolbox where we can easily reach them? Or should we bury them and only dig them out again in the event of a severe crisis? And should we redefine our target?
The debate on these issues, in academic and policy circles alike, has been lively. So let us weigh the potential costs and benefits of redefining our toolbox and our strategy.
Let’s start from the beginning. The crisis called for very loose monetary policy, and we responded with our standard tools.
However, despite the ECB’s efforts, inflation remained at historically low levels. In January 2015, the Eurosystem thus announced that it would start buying government bonds. A year and a half later, in June 2016, it also began to buy corporate bonds.
So, the basic story is simple. Once the Eurosystem had come to the conclusion that the standard tools were not sufficient to reach our objective, it turned to non-standard tools.
Asset purchases are one of those tools. And they have made our job as central bankers even more complex than it used to be. The first challenge we faced was how to go about implementing the purchase programme. Interventions like this can have any number of unintended consequences.
First, as a major actor in the markets for public and private bonds, we have to tread carefully. We have to make sure that the markets continue to work well. We thus placed limits on the share of a bond issue that we would buy, for instance. This ensures that the Eurosystem does not form a blocking minority in the event of a debt restructuring involving collective action clauses.
Second, there is a risk that we might become a drag on market liquidity, in particular in some segments of the corporate bond markets. We have thus tried to minimise this kind of side effect, and not to worsen existing shortages of certain assets.
Third, we must be aware that our actions might distort prices. The more bonds we buy and the longer we go on buying them, the greater the risk that prices no longer reflect market conditions.
These are some of the risks we have to bear in mind. Viewed from today, however, it seems that our non-standard measures did have the desired effects on financial markets and, consequently, on the real economy.
But what does this mean for the future? Seeing the short-term effects, some observers suggest making the balance sheet part of the new orthodoxy of monetary policy strategy – or part of the standard toolbox, as they like to say.
Let me say clearly that I think this should not happen. The short-term benefits should not lead us to overlook the long-term costs and the severe side effects. These costs and side effects can be quite high, and they tend to grow over time. Thus, a comprehensive asset purchase programme, such as the current one, should only be used to fight a real risk of deflation. In my view, that is the only valid reason for using such tools.
For one thing, I wonder how it would affect the incentives for various stakeholders. Financial market participants, firms, consumers and also governments have to varying degrees adjusted to historically low interest rates and ample liquidity. And for some of them, low interest rates have certainly made life easier. It is therefore crucial to realise that this won’t go on forever – the exit will come eventually – and to adjust accordingly.
Second, there is the issue of financial stability. When the central bank buys bonds, it might also fuel asset price bubbles, financed by cheap credit. This risk increases the longer unconventional monetary policy measures continue to be applied.
And third, monetary policy can only buy time – no matter how loose or unconventional it becomes. Monetary policy is not a tool meant to alter the long-term viability of economies, and it would be dangerous to believe otherwise. I will come back to this point in a few minutes.
And it’s not just about the tools, it’s also about the target. A second idea that is often put forward is that we should change our objective and strive for higher inflation.2 Proponents of this idea claim that we should no longer aim for inflation rates of below, but close to, 2% over the medium term. They argue that we should aim for inflation rates higher than that. This idea is based on the observation that higher inflation rates generally lead to higher average nominal interest rates. This would then give central banks more room for manoeuvre, as they would be less likely to hit the effective lower bound.
In my view, aiming for a higher inflation rate would lead to the same problem as making non-standard tools a standard part of the toolbox. Such proposals tend to underestimate the associated costs.
There is evidence that changing the objective would be too blunt a move. Hitting the effective lower bound is costly, but so is higher inflation. Research shows that the key benefit of higher inflation – fewer and less severe zero lower bound events – is easily offset by the costs of positive trend inflation.
What are these costs? Well, price distortions become larger, overall inflation becomes more volatile, and the welfare costs of such volatility increase as well.3
And changing our objective could undermine the markets’ confidence in the ECB’s commitment to price stability. The transition could prove very hard to manage and could take many years. And during that time uncertainty would increase. We would thus likely witness increased disagreement among forecasters – inflation expectations might become more volatile. And this is anathema to central bankers. In the end, it is part of our job description to ensure that long-term inflation expectations are well anchored.
In this context, we must also take into account that changing our objective might in itself hurt our credibility – the main asset of any central bank. And this would make it much harder to steer inflation expectations even after the new objective had been met. After all, people might think that at some point in the future, we could revise our objective again. Central bankers are conservative for a reason.
To sum up, it seems to me that these proposals have one thing in common. They are a reaction to the dramatic shifts we have witnessed since the crisis. Hitting the effective lower bound forced a bout of creativity in the world of monetary policy. But I do not believe that the way forward is to entirely revolutionise the way we think about the role of central banks.
Still, we should also be realistic. The crisis revealed shortcomings in our thinking, and this is something we need to work on.
What about financial stability?
So the debate about the future does not end here. Another thing to discuss is how monetary policy and financial stability interact. And this debate also involves our standard tool, the interest rates. Interest rates have been very low for quite some time now. This has coincided with an upsurge in asset prices and a compression of spreads on risky securities.
Against this backdrop, some economists argue that monetary policy has an effect on both the capacity and the appetite of investors to take on risk.4 This has been termed the risk-taking channel, and it does not play a role in standard monetary policy frameworks. These focus instead on the interest rate, credit or bank lending channels.
Now, how does this risk-taking channel work? Well, it can work in different ways. For a start, low interest rates mean low returns. Investors might rebalance their portfolios and take on more risk in order to make up for lost returns. Should we be concerned about this? Well, if the process is well managed, there is no need for concern.
However, some financial institutions set rather inflexible target rates of return, for example because investors expect a certain minimum yield. These inflexible targets may lead them to take on more risk than appropriate. It is clear that this could lead to trouble further down the road.
Also, the financial system’s behaviour can be procyclical. There is reliable empirical evidence that loose monetary policy lowers both the measured risk in asset prices and the risk premia demanded by investors.5 But value-at-risk approaches mean that the risk management of banks relies on measured risks. This could then lead to increased leverage, which would add even more liquidity to the global financial system. So there is a view that the risk-taking channel should be incorporated into the models that guide monetary policy.
Let’s try to unpick this argument. In my view, there are two conceptual issues here. The first is the idea that policymakers should only be concerned with risk-taking if it is excessive. But establishing whether or not risk-taking is excessive is quite hard. The second issue is that even if we conclude that there is excessive risk-taking, it is far from clear whether monetary policy would be the best tool to deal with it.
Let’s start with the first issue. What qualifies as excessive risk-taking? Well, there are two conditions. Risks would need to be mispriced, and investments would need to have been financed with a dangerous amount of debt. If these two conditions are met, a financial crisis becomes more likely.
Now, where do we stand on these two issues? Well, there is some evidence that banks in the euro area do indeed provide riskier corporate loans when short-term interest rates are low.6 It seems that banks tend to grant loans with fewer covenants and lower collateral cover. Such loans are riskier in the sense that in case of a default, the impact on the banks might be larger.
However, it is hard to tell whether this kind of risk-taking is excessive or not. Many studies rely on credit register data which do not include loan interest rates. Thus, it is hard to tell whether the loans are mispriced given the risk they contain. It could just as well be that banks grant loans to riskier companies but do so at higher interest rates, which compensate fairly for the higher loss risk.
This brings us to the second issue. Even if we could say with certainty that risk-taking is excessive, we would still need to be sure that monetary policy is the right tool to deal with it.
Some think it is; others think it isn’t. The latter believe that central banks would run into trouble if they tried to deal with financial stability. This is because it would involve a conflict of interest. To stop a credit-induced bubble, we would need to raise interest rates, and we might need to raise them by a lot. After all, potential profits are huge when there is a bubble, so investors would not be deterred by a modest increase in interest rates. But what if a huge increase in interest rates would derail inflation? Which path should monetary policy take in such a situation?
As the economist Jan Tinbergen has observed, you cannot achieve two goals with just one tool. And attempting to do so would damage the credibility of central banks.
It’s not easy to say who is right. Financial crises are fortunately rare events, and usually there is no single cause. This means, however, that we do not have a whole lot of data to analyse. Still, in the past, monetary policy was used at least in part to prevent bubbles from forming. This was done in the United States in the late 1920s, for instance, and in Japan in the late 1980s. Suffice to say, these episodes did not end well.
So, it seems wise not to rely on monetary policy when it comes to financial stability. And in this sense, the reforms of the institutional architecture in the Eurosystem after the crisis were indeed wise. The ECB’s mandate remains clear: our objective is price stability. But we have also been given powers for microprudential supervision, as well as some macroprudential competences.
On the microprudential side, European banking supervision contributes to the soundness of the banking system. A lot of work has already been done, and we will continue to do more.
First of all, banks are much better capitalised and provisioned than they were in 2014, which makes them more resilient to shocks. Second, we are pushing banks to reduce legacy assets such as non-performing loans. And third, we are requiring banks to improve their risk management, for instance with regard to the internal models they use to calculate the risks associated with the assets they hold.
Macroprudential policy was also developed in the aftermath of the crisis. It still needs to be refined and tested. Available evidence from the euro area and beyond may not be definitive, but it is encouraging. Let me give you a few examples.
With the 2010 Basel III agreement, we introduced countercyclical capital buffers. These can help lean against the build-up of imbalances, and if they cannot fully prevent them, they can at least make banks more resilient in the event of a downturn.
Borrower-based tools also look promising. In Asia, they have been used for many years now. In South Korea, loan-to-value and loan-to-income limits have successfully curbed growth in house prices.7 But again, more systematic evidence is not clear-cut. Studies based on cross-country evidence show that only loan-to-income limits have had a discernible effect on house prices. But there is still reason to be cautiously optimistic.
And the benefit of macroprudential policy remains clear: it is the most direct way to deal with financial stability concerns. And, over time, we will gather more evidence and gain more experience. This will give us plenty of opportunity to fine-tune these new tools.
Beyond monetary policy: wealth, income and structural reforms
But the discussion about the wider effects of monetary policy is not just about financial stability. Particularly here in Germany, there is a fierce debate about who loses out as a result of monetary policy. Many savers feel that they have to bear a huge burden because interest rates are so low. I understand their frustration. After all, I am a saver myself.
However, we are not just savers. Some of us might have taken out a loan to buy an apartment or start a business – we benefit from low interest rates. Some of us may have invested in stocks or bonds – we benefit from rising markets. And some of us may have kept our jobs because monetary policy propped up the economy.
Monetary policy affects people in different ways. In more technical terms: it has an impact on the distribution of wealth and income. And the important question is whether it makes this distribution more unequal. Does it make the rich richer and the poor poorer?
Let’s look at wealth first. Here, the most important item for many households is real estate, the home they live in and any other property they may own. Real estate makes up about 80% of the total assets of households, almost irrespective of their level of prosperity. So, to answer our question, we need to think about how monetary policy affects house prices.
The answer is that quantitative easing raises house prices and, on average, this boosts the net wealth of all households, rich and poor alike. In fact, poorer households might even benefit more because they are more highly leveraged. So their main asset – their house – gains in value, while the debt they used to finance it becomes cheaper.
Now, what about incomes? Well, loose monetary policy props up the economy and thus raises employment. And again, this should mean that people with lower incomes benefit most. After all, unemployment is much higher among those with low incomes than among those with high incomes.
So, it seems that loose monetary policy might also reduce income inequality. However, there are three things that we must keep in mind.
First, the effect of monetary policy on inequality is quite small. In the end, inequality is driven by many things such as technological progress, globalisation and taxes. Compared with these things, monetary policy seems to play a minor role. Second, the effect is temporary. In the long run, monetary policy is neutral; it does not affect real variables such as output. And third, the effect is symmetric. Those who benefit when monetary policy is loosened suffer when it is tightened.
We have to be aware of these effects, of course. But should it be up to monetary policy to actively engage in redistribution policy? Well, monetary policy has a very clear goal: price stability. So, I would argue that monetary policy should focus on this goal and leave other goals to other policy areas, and that includes redistribution policy.
Imagine inflation were sky-high, for a change. Should monetary policy then refrain from raising interest rates because it might alter the distribution of income and wealth? Again, there is a clear conflict of interest as one tool would have to achieve two objectives at once.
That said, the topic of wealth and income distribution is an important one and it should not be taken lightly. I therefore welcome the fact that academia is paying increasing attention to the matter. The ECB is contributing to the debate with our household finance and consumption survey.
And this links to a broader issue. In the crisis, monetary policy has played a key role. However, it is not the only game in town – it cannot be the only game, and it should not be. Monetary policy is not a substitute for other policy areas such as social policy, fiscal policy or labour policy.
As I and many others have repeatedly said, monetary policy cannot resolve the structural problems of the euro area. All it can do is buy time – at an increasing price. So it is time for politics to kick into gear, overcome reform fatigue and do what needs to be done. The current conditions are very favourable.
The good news is that many countries have come a long way. Ireland was the first success story, and I am glad to be able to add countries such as Spain and Portugal to the list. These and other examples show that structural reforms pay off and pave the way for growth to return.
But coming back to inequality, there is one thing we must do. We must ensure that economic growth benefits everyone. To that end, we should not just rely on redistribution policy. It is rather about creating equal opportunities for everyone. In this context, education and equal access to it should be at the top of the to-do list of any government.
So, a lot remains to be done, not just on equality but on growth in general. After all, growth also plays a big role in fighting inequality.
And when it comes to long-term growth, research shows that the quality of institutions plays a major role. 8 At the same time, the quality of institutions differs between countries. So, some countries have scope to improve judicial systems, public administration and insolvency regimes. This would make their economies more resilient and help them to grow.
And I’m not just referring to institutions at the national level. European institutions are not free of deficiencies either. They too need to improve. If we want European citizens to trust Europe, we need to make sure we deliver on our promises.
And there is a long way to go: according to the latest Eurobarometer survey, only 41% of Europeans tend to trust the EU. And the latest election results show that anti-European parties are still on the rise – think of Germany, think of Italy. So there is a need for reform and a need to explain thoroughly how the European Union benefits each and every European citizen.
Conclusion
Ladies and gentlemen,
History does not end for monetary policy. It did not end with the great moderation, and it will not end with the exit from our unconventional measures. There are still many questions that need to be answered, some of which I have discussed in my speech.
But although monetary policy still needs to evolve, it must do so within a certain framework. And this framework is defined by two things.
First, even though central banks have assumed a key role in the crisis, their objective is very clearly defined: to ensure price stability.
Second, to achieve this objective, central banks need to be independent.
Within this framework, we need to answer all the questions which remain open. So, there is still a lot of work to be done in terms of both theoretical and empirical research. That’s where universities come into play. They can help us to gain additional insight, and they can help us to come up with new ideas. From an academic viewpoint, these are certainly interesting times.
Now, history might not end, but you’ll be glad to hear that speeches do. I have raised a few issues that are currently being discussed in connection with monetary policy. And now I look forward to having a discussion with you on these and any other issues you might want to raise.
Thank you for your attention.
1 European Economic Forecast, European Commission, Autumn 2017.
2 See Blanchard, O., Dell’Ariccia, G. and Mauro, P. (2010), “Rethinking Macroeconomic Policy”, Journal of Money, Credit and Banking, 42, 199–215; and Krugman, P. (2014), “Inflation targets reconsidered”, ECB Forum on Central Banking, May 2014.
3 Coibion, O., Gorodnichenko, Y. and Wieland, J. “The Optimal Inflation Rate in New Keynesian Models: Should Central Banks Raise Their Inflation Targets in Light of the Zero Lower Bound?”, The Review of Economic Studies, Volume 79, Issue 4, 1 October 2012, pp. 1371–1406.
4 See, for example, Borio, C. and Zhu, H. (2008), “Capital regulation, risk-taking and monetary policy: a missing link in the transmission mechanism?”, BIS Working Papers, No 268, BIS, and Adrian, T. and Shin, H.-S. (2010), “Liquidity and leverage”, Journal of Financial Intermediation, Vol. 19, pp. 418–437.
5 See, for example, Bekaert, G. , Hoerova, M. and Lo Duca, M. (2013), “Risk, uncertainty and monetary policy”, Journal of Monetary Economics, Vol. 60, Issue 7, pp. 771–788, and Heider, F., Saidi, F. and Schepens, G. (2017), “Life below zero: Bank lending under negative policy rates”, mimeo, available on SSRN.
6 See Maddaloni, A. and Peydro, J.-L. (2011), “Bank Risk-Taking, Securitization, Supervision, and Low Interest Rates: Evidence from the Euro Area and U.S. Lending Standards”, Review of Financial Studies, Vol. 24, Issue 6, pp. 2121–65 and Jimenez, G., Ongena, S., Peydro, J.-L. and Saurina, J. (2014), “Hazardous Times for Monetary Policy: What Do Twenty-Three Million Bank Loans Say About the Effects of Monetary Policy on Credit Risk-Taking?”, Econometrica, Vol. 82, Issue 2, pp. 463–505.
7 Igan, D. and Kang, H. (2011), “Do Loan-to-value and Debt-to-Income Limits Work? Evidence from Korea”, IMF Working Paper, IMF.
8 Acemoglu, D., Johnson, S. and Robinson, J. A. (2005), “Institutions as a Fundamental Cause of Long-Run Growth”, in Aghion, P. and Durlauf, S. (eds.), Handbook of Economic Growth, 1st edition, pp. 385–472.