Concluding remarks by Director Luís Laginha de Sousa at the Joint ECB and Banco de Portugal Conference on "The use of narrative information in assessing the effectiveness of macroprudential policies"
It is for me a great a pleasure and an honour to come here before you and to close this Conference which was jointly organized by the ECB and the Banco de Portugal.
In the coming minutes I would like to share a few thoughts as a wrap up to this Conference.
But before starting to do so, I would like to express a very deep thank you, both to the speakers, for their contributions and also to all other participants, for the lively and thorough debates we’ve had all day.
I believe that the twelve presentations, have greatly contributed to shed light on a number of issues which are extremely important for the implementation of macroprudential policy.
Before moving into the “thoughts” component, just two preliminary notes.
The first note is to say that this Conference could not come at a more appropriate time, not just for the international audience it caters to, but also for the Portuguese audience, given the very important topic currently being discussed, concerning the future of the supervision architecture in our country.
For the avoidance of doubt, I must add that having this Conference here, in the middle of such a relevant discussion, was a very fortunate coincidence. I’ve used the expression “fortunate coincidence” because initially the Conference was planned to take place in “The Netherlands”, but some months ago, the ECB approached the Bank of Portugal to host this Conference in Lisbon and at that moment, there was no perception that it would coincide with the debate about the supervision architecture, at national level.
The second preliminary note is to say that when it comes to effective communication there are some fundamental rules, of which I will try, as much as possible, to apply two of them. One rule is to focus on few but important messages. The other rule is to draw simple conclusions.
Having this in mind, what I will try to do is to link some of the takeaways from this Conference, with key components of the already mentioned debate about the national supervision architecture.
To start with, it is adequate to briefly point to “Why it is important to have macroprudential policy” and, which are the key supporting factors, to make it successful. To do so, particularly to answer to the “why”, we have to go back a few years in time.
In the aftermath of the international financial crisis came the consensus that risks related to excessive leverage, the combined effect of financial institutions’ behaviour and signs of speculative bubbles had been largely neglected in the preceding period, therefore, the safeguard of financial stability required the implementation of an autonomous policy (and I have to stress the word “autonomous”) aimed at mitigating systemic risk.
The urgency in operationalizing this policy led many countries to the development of a toolkit, well before a comprehensive theoretical framework and empirical evidence on the impact of these measures, was stabilized.
Issues like the complementarity and substitutability between several macroprudential policy instruments, their interaction with other policies, their role when risks have already materialized and the channels for possible spill-overs, still need to be fully explored.
The research that was presented in this conference is a huge step in this direction.
The databases used in many papers, with aggregate and cross-country comparable information on the activation and on the release of macroprudential policy instruments, are truly a valuable tool for research and policy.
This narrative information provides greater depth and coverage to the empirical studies on the effectiveness of macroprudential measures, which is of great importance both for researchers, analysts and policymakers.
Information and rigorous analysis is key for decision-making (rather than feeling and prejudice).
So from what we heard today, the effectiveness of macroprudential policy depends on the right combination of timely action with informed decisions.
Timing is particularly relevant since the preventive nature of the policy implies that it should act before risks materialize and become too difficult to manage.
Informed decisions is also a key component, which goes hand in hand with timely action.
If we were to use a culinary metaphor, we could say that it is a little bit like baking a cake, the eggs and the flour have to be combined not just in the right quantity but also at the right moment. I apologize if my culinary knowledge failed to produce the right metaphor, but I hope that at least the idea was there.
There are a number of features inherent to “macroprudential policy” which make it a very special kind of creature.
In particular, I would like to draw your attention to three of the features, which have also been addressed either directly or indirectly in today’s Conference.
The first of these features, which is strictly linked with the timing component mentioned before, is the fact that macroprudential policy can be subject to what is described as “inaction bias”.
There are several factors which can contribute to limiting the action of macroprudential authorities in face of vulnerabilities in financial systems.
Contrary to what is normal in the corporate business world, where success has many parents and failure is an orphan, when it comes to macroprudential policy, it is not quite so. Success is orphan and, when failure materializes, there are too many parents, but none seems to be willing to accept the parental responsibility, unless it is forced to do so.
So, the preventive nature of macroprudential policy implies the activation of instruments at a stage where risks have not yet materialised and because of that, the risks at such point are not perceptible for economic agents.
The efficacy of the policy in preventing a crisis cannot be verified either, and that’s because a crisis which has not yet occurred does not leave a trace.
The fact that the financial cycle is typically longer than the business cycle makes the lag between policy benefits and its costs, wider for macroprudential policy, than for most other policies.
Moreover, not having a clear and measurable objective (as there is for monetary policy) makes it more difficult to assess its effectiveness.
So, in short, while the benefits of macroprudential policy may be uncertain and delayed, its costs are more observable and immediate. As such, macroprudential policy measures, are likely to be unpopular.
Imposing tighter requirements when risks might not be apparent may also conflict with the objectives of other policies. Or, to put it in a more proverbial way, “it’s like taking the punch bowl away while the party is still going”.
The second feature inherent to Macroprudential policy is its potential overlapping or even contradiction with monetary policy.
Assessing the effectiveness of macroprudential policy implies a thorough understanding of its transmission mechanism and of all the factors which might affect its policy targets.
As we have seen in the papers presented today, disentangling the effects of macroprudential and monetary policy can be quite challenging, not just because of the interdependency between the two policies, but also because of their impact on credit growth and asset prices.
The liquidity conditions of an economy and their impact on aggregate demand, which are essentially of the responsibility of monetary policy, are determinant to the financial cycle.
Loose monetary policy can promote the emergency of speculative bubbles and also what is usually described as the ‘search for yield behaviour’.
Therefore, the orientation of monetary policy is not irrelevant to the achievement of macroprudential objectives.
On the other way, macroprudential policy, by affecting credit conditions and the resilience of the financial system, can impact the transmission mechanism of monetary policy.
Understanding and managing these interactions is crucial for effective policy implementation.
The third and last feature of macroprudential policy I would like to mention in my final remarks, is the requirement of a strong analytical knowledge on systemic risks and financial markets, in order to achieve an effective policy implementation.
I believe this a self-explanatory feature and therefore I will not elaborate on it.
Having in mind the three features I’ve talked about, probably a fundamental question to be raised is how to take such features into account to ensure that the right policies will be implemented when they need to be implemented.
And I believe the answer to that question can be summarized into a very simple two words sentence: “Good Governance”.
Without good governance it is impossible to guarantee that the right measures will be taken when they need to be taken, and when it comes to good governance in macroprudential policy we are talking about three basic principles that have to be present at all times (and the order does not necessarily reflect hierarchy of importance): Accountability; Credibility; and Independence.
Coming to this point of my final remarks and not drawing conclusions on what I’ve said before, would be a little bit like not to talk about the “elephant in the room”. So to conclude, it is not irrelevant which authority should be responsible to implement macroprudential policy.
If we carefully and openly assess the conditions which have to be met to ensure good governance, it won’t be difficult to recognize that central banks are, in general terms, better positioned to do so.
They have the capacity to analyse systemic risks and to act in a timely way in preventing and mitigating such risks, independently of lack of public support or conflicts with other policies.
They have the competence in the analysis of macroeconomic and financial developments, and the expertise in assessing risks to financial stability.
They have the experience in the implementation of monetary policy, which is particularly relevant given the interlink between such policy and the macroprudential one.
They are those enjoying a greater degree of independence as it is enshrined in their Statutes and in the Treaties of the European Union as an essential condition for the fulfilment of their mandate of ensuring price stability.
It is, nevertheless, important that, at the internal level of the institution, there is a clear separation of some functions, in order to ensure autonomy and prevent conflict of interests in the pursuit of different goals. But that does not hinders all that was previously said.
With this remarks, which have a strong “local flavour” but totally in sync with the different flavours of the topics covered during the day, I would like to conclude this conference.
I hope you have not only enjoyed but also found this conference useful for the challenges that lie ahead.
I wish you all a great weekend and a safe return home.
And thank you for being here and for having made this conference possible and a success.