Agustín Carstens: ‘the Basel regime has been very good for the banks’

Good morning. It was interesting that so much of the Wall Street commentary on Wednesday’s Fed meeting declared it to have been “dovish”. And it is true that Jay Powell pushed back on the possibility of a rate rise. But he also emphasised that he is prepared to be extremely patient about cutting. What’s really going to be dovish or hawkish is the data. Email me: robert.armstrong@ft.com

Friday interview: Agustín Carstens

Agustín Carstens has been the general manager of the Bank for International Settlements since 2017. Previously, he was governor of the Bank of Mexico, and on the board of the IMF. The BIS serves as the “central bank for central banks”, providing a forum for co-operation and assisting national central banks in managing their foreign exchange reserves. It is also the host of the Basel Committee on Banking Supervision, the pre-eminent global standard setter for the banking industry. Unhedged visited him in his office in Basel last week, where the conversation ranged across bank failures, capital requirements, private credit and the “Finternet”.

Unhedged: The Basel committee is an international advisory body with no treaty power. Its recommendations are just that — recommendations. A bunch of people come together in a room in a mid-sized city in Switzerland and decide what global banking rules should be, and to some degree, the world does what the committee recommends. How is this possible? It looks a little like magic.

Carstens: Whatever happens here at the BIS and whatever happens in the Basel committee, there is a lot of hands-on involvement by [central bank] governors. Compare, for example, the IMF. It has an in-house executive board, and board members — I have been a board member — always have to consult with their capital before they make a decision. Here, governors come, they discuss, they get their hands dirty, and at the end of the day most of the time the important decisions are made by the top leadership in the room. There is a lot of ownership. 

The other thing is the mechanics of how decisions are struck. The fact that governors are very aware of the state of their banking system allows them to know, is this going to fly or is this not going to fly? 

Finally, I think that the Basel principles are a very useful construct for the banking industry. I mean, you hear a lot of complaints about Basel III. But it has become a tremendously effective mechanism to provide signals about how strong an institution is. The fact that you are able to say, ‘I’m Basel III compliant,’ or ‘I am above the minimum standard by so much,’ that provides a lot of information. That has helped a lot of institutions. It has helped the whole system. 

There is always some stress and friction, but I think that friction has led to a reasonable balance, where we as authorities have a sense of how strong an institution is, and the banks have some measure of where they are, and the market takes that into account. Fundamentally, the Basel principles address the right issues. They ring the bells of common sense about what central bankers think about: liquidity, leverage, capital.

Unhedged: You mention friction. Bankers have complained there are 14 different Basel endgames and there is fragmentation at the implementation level. Do you think this is an unsolvable problem, given the many countries and interests involved?

Carstens: Fragmentation is a very strong word. Differentiation is a more accurate description. In a fragmented system, you cannot talk to each other, or there is no engagement between systems. Fragmentation would mean that you could not operate as an integrated entity in different countries. Yes, there is deep differentiation, and the differentiation is there by design because the basic principles establish minimums. The rule is not that you have to comply with this particular figure; you have to at least observe the minimum.

Unhedged: UBS had its annual meeting this week. The headline in the Financial Times was the bank is concerned about high Swiss capital requirements. And we have heard similar complaints from American bankers. How does the BIS respond to the idea that capital requirements have gone too far?

Carstens: I go back to what I said. The Basel regime has been very good for the banks, for the system overall. Today the Gsifis (Global Systemically Important Financial Institutions) are strong, and part of that is because they have observed the standards very well. Since the global financial crisis, they have made tremendous progress. 

Unhedged: A more general question about central banking. When central bankers are asked a hard question — about the effect of zero rates, or it’s about the climate, or some other global issue — they tend to reply, “I’ve got a simple mandate, price stability and full employment.” The consequences of following that mandate are not the banker’s problem. But one might look at the long period of very low rates and the consequences that has had for the economy, for example, and wonder if such a tight mandate is optimal.

Carstens: For me, the question has a relatively simple answer most of the time, if not all the time. A mandate by law is accompanied and by a set of instruments designed to allow the compliance of that mandate. If you want autonomy to work, you need to give the central bank a clear mandate and the instruments to comply with it. And you have to have an accountability regime, where they can explain what they are doing towards the mandate. 

There is the perception that if you command money, you can do many things, and there is always an endless list of things being suggested that the central bank should do. Once you start deviating from the mandate, you enter into territory where you might have to do a trade-off against your mandate, for the simple reason that you might not have been granted the right instrument. And of course no central banker wants to be put in that position.

Unhedged: I’m trying to ask a higher-level question about institutional design, from the point of view of good government. There is a way that central bank autonomy might be too effective. Because the mandate is clear and the bank is autonomous, and the rest of government is kind of messy, central bank policy gets priority, in a way, and maybe it shouldn’t. 

Carstens: Well, let me say the following. I think the whole issue of the mandate needs to echo society’s preferences. Now, one thing that has happened is that once inflation came down in the 1970s, to some extent we forgot about it. That’s the state of nature. But the cost of that is that people started not appreciating the value of keeping inflation low and stable. Why should you be concerned about something that doesn’t exist? And suddenly inflation pops up and now you see it’s a huge issue. I think if you ask people now, do you think it’s important to have an institution within the state whose primary task is to keep inflation under control? I would say that there would be a lot of political and social support, because people have seen the impact of inflation very recently. And that’s what grounds central banks.

Unhedged: Is there a danger that we are not back to a low inflation world? We’re much closer to target than we were, but is it possible that central banks have not done what’s necessary?

Carstens: I think so far things are working quite well. We had a unique inflationary shock. It had the combination of tremendous supply shocks with a very important aggregate demand push, through monetary instruments and fiscal instruments. To bring that inflation under control required very forceful and decisive action. And I think central banks have acted [appropriately]. Now, we warned about the last mile many, many months ago. This is not new. The fact that the last mile will be bumpy is not necessarily news; even more so given that we have had tremendously traumatic events as part of the origin of this inflationary shock. We have never had a situation where the world economy was put under suspended animation to solve a health crisis. That is huge, and 100 per cent new. Did we think that it was going to work like when you turn the lights on and off? The economy doesn’t work that way. If you turn off the economy, different sectors respond in different ways. Supply chains get completely distorted, the labour market gets completely distorted. The government had to respond in a very aggressive way. So for the economy to find its way again will take time. We are in that process; we have to be patient. 

At the end of the day, the last mile is about allowing for the needed relative price adjustments, and there are still many relative price adjustments to be made, and there are still doubts about whether some relative prices will not come back and bite. We know very well that monetary policy acts with a lag. But I see a strong commitment from central banks, and I think they should continue to be persistent. I think we are very close, and I think it is very important to go back to the 2 per cent inflation target.

Unhedged: Last year we had some major banking incidents, one here in Switzerland, several in the United States. What did we learn?

Carstens: In the case of Credit Suisse, it was clear that business plans really make a difference, and business plans are very difficult to assess from a supervisory point of view. Probably the supervisory authorities should have had more strength in being able to influence Credit Suisse, because if you talk to the relevant authorities, the dialogue between them and Credit Suisse was there. But that dialogue didn’t carry enough strength to make Credit Suisse react in time. So I think that giving more [power of] persuasion to the supervisors/regulators would have helped, including sanctions and so on. 

Coming from Latin America, the Credit Suisse saga was like García Márquez’s A Chronicle of a Death Foretold. You could see it coming. I think authorities saw the risks, and they pushed and they pushed and they pushed, but they didn’t get to where they wanted to go, and it took a huge world shock to accelerate the process. So reflecting on the events in the US banks and here in Switzerland last year, I think we authorities should focus more on governance, risk management and business models. Authorities need to have a way to have a more direct impact on the institution. This is not so much in the field of Basel principles, but about how relationships between regulated entities and authorities work, and how at some point, you have to be able to call the bank, have a discussion, and try to steer the bank in the right direction. 

Unhedged: Some observers have expressed the concern that, in the case of Credit Suisse, there is not enough clarity about how the resolution process was supposed to work, and that therefore the resolution couldn’t go as expected or got stuck, leading to the UBS takeover. Do you agree?

Carstens: No, I think the resolution work was there. I think it could have been done. But resolution is an option, not the only option. As an authority, you have to evaluate how it will be less painful. And the way it was engineered, they used many of the principles of resolution. They stabilised the situation effectively. The Swiss taxpayer didn’t lose money. The world economy, the world financial market, did not get traumatised.

Now, going to the US regional banks, there the issue was really very elementary banking mistakes. I think it is important to mention that many of the Basel III principles did not apply to those banks. With the benefit of hindsight, if they were subject to those principles, probably accidents could have been mitigated. And that is being remedied now.

Unhedged: That’s because they would have had more capital and higher liquidity?

Carstens: Yes, and in the case of a Silicon Valley Bank, the marking to market of their positions would have been passed to equity early on. And when you have to do that constantly, the market takes notice early on. Here, once they went out and said to the market that they needed capital, the market said, why do you need capital? Well, we have all these losses that you don’t know about. That is when everything exploded. 

Unhedged: Changing topics, I agree with many of the goals you articulated in your paper about the Finternet — using unified ledgers and tokenisation to reduce the frictions that persist in the global financial system. But I wonder if a more interconnected, frictionless system might increase cyber security risks — if the current patchwork has some security advantages. 

Carstens: With the Finternet, we’re not talking about a monolithic network or a monolithic ledger. We talk about a network of networks. An advantage of this rethinking of the financial system is that you can incorporate these security concerns into the design. One problem we have today is that many systems, not all of them, were not necessarily designed with cyber security in mind, and I think it makes a huge difference if you can incorporate everything from the outset. Cyber security will be an endless battle. But I think that security would be enhanced when you have it in the design from the beginning.

Unhedged: What’s the next step in realising your vision?

Carstens: One of our projects, Agorá, is an effort that is being put together by the BIS and the IFF [the Institute of International Finance]. The BIS are coordinating with seven very important central banks — the Fed, the Bank of England, the ECB, Swiss National Bank, the Bank of Korea, Bank of Japan and the Banco de México. And then the IFF is putting together a group of financial institutions. If you think about the Finternet, a key component would be the circulatory system — the payment links. The payment links will depend on wholesale CBDC [central bank digital currencies] and tokenised deposits. Tokenised deposits is a very unfortunate name. It is better to talk about digital commercial bank money. So, in the Finternet, the payment system would rely on wholesale CBDC, which is very similar to the reserve system we have today, but empowered by programmability and so on; and on commercial bank money, which is exactly what we have today in two-tier systems. So Agorá is going to study all of the consequences of having tokenised central bank money and tokenised commercial bank money, with a very specific application to correspondent banking. Even if we don’t [progress all the way] to the Finternet, this has a lot of value. Because correspondent banking is a system that, by and large, needs huge amounts of fixing. 

Unhedged: Does it concern you — and I don’t want to sound too cynical here — that a lot of banks involved in correspondent banking are, as it were, in the friction business? What we think of as inefficiency, they think of as fee income.

Carstens: A lot of the reasons why correspondent banking is a pain is because it’s very complex, and it’s based on very old technology. A lot of the costs have to do with compliance, have to do with AML-CFT [anti money laundering, combating financing of terrorism] requirements. It has to do with different rules in different places and all that can be streamlined. Many banks have left correspondent banking because it was perceived as a low-margin, high-risk enterprise. And I think with the Finternet you can improve that balance. You can dramatically reduce the risk for the institution. 

Unhedged: There’s been two big changes in the banking world recently. One is we went from a zero rate world to what we might call normal rates. Another big change is that sovereign debt levels are much higher. How will those two big changes affect the work that you do and the work that central bankers do?

Carstens: If — and this is a big if — we settle at a rate that is higher than what we had last decade, I think that’s good for central banks, because it creates some distance with the zero lower bound, and gives more flexibility to monetary policy. And I think having rates back to normal shouldn’t be such a detriment to growth. So I think that’s fine. 

I am concerned about debt over GDP in many countries. And it’s not only not only high debt, it’s also high deficits. It is not very good to have fiscal and monetary policy at cross-currents. And to some extent that is happening. So the work of bringing down inflation becomes more difficult at the margin. And with high debt over GDP, you know the markets will have to carry a larger stock of the “safe asset” [sovereign bonds]. But by how much can you increase the stock of safe assets before their quality, safety, is questioned? If you come from an emerging market, you know that can happen. 

Unhedged: Assets have been rushing to private credit. How do you assess those risks associated with non-bank lending?

Carstens: The Basel committee and the FSB [Financial Stability Board] are looking primarily into the connection of non-bank financial activities with banking activities. Because at this stage, that could the transmission mechanism. There are two important aspects there. One is whenever you make a rule, there will be efforts to try to avoid it. If a lot of the activity is that going on [in non-bank lending] is regulatory arbitrage, that probably is not good for the system overall. And the other important consideration is that we haven’t seen how [new activities in the non-bank financial world] will perform in the full interest rate cycle. We don’t have enough information to say if they are handling the cycle well or not. The banking system, we know how it is going. We don’t know exactly what the exposures are for different sectors, what the interest rate sensitivities are, and so on in non-banking activities. There will be casualties. And we don’t know how high they will be. At least we as authorities should try to have much better information. I’m sure that there are many non-bank activities that are worthwhile and that are safe. But there might be many others that are not. 

One good read

Brittney Griner talks about her imprisonment in Russia.

Source: Economy - ft.com