Goodharts Law, named after Charles Goodhart, the British macroeconomist and erstwhile member of the Bank of Englands Monetary Policy Committee, was first formulated in 1975. It says: "Once a social or economic indicator or other surrogate measure is made a target for the purpose of conducting social or economic policy, then it will lose the information content that would qualify it to play such a role."
Though this sounds abstract, it is easy enough to think of examples of Goodharts Law in action and the unintended consequences it wreaks as a result. If food manufacturers are told to cut the amount of saturated fats in their products to below a specified level, they might instead use cheaper and even more dangerous trans fats. If pubs are told to ban happy hours, they might replace them with two-for-one offers.
By codifying a group of 28 globally systemically important banks (G-SIBs) and compelling these entities to hold further common equity tier one capital on top of the 7% (4.5% plus the 2.5% capital conservation buffer) minimum demanded by the Basel Committee on Banking Supervision by 2016, the Financial Stability Board (FSB) has stated a single overarching policy goal: "The creation of a safer and sounder banking and financial system." Goodharts law suggests that may prove
What the FSB does not say is how its aim will be achieved. But defining the G-SIBs is only part of the solution. The banks that fall outside the club will play an equally important role. By shouldering systemically important banks with higher capital requirements, one aim of policymakers seems to be that smaller banks will gain market share from their larger G-SIB rivals. This, regulators hope, will offset some of the risk to the economy and implicit moral hazard of creating the G-SIB club by increasing competition.For this aim to be realised access to funding for the non-G-SIB banks is critical. But many market participants are convinced that the debt of non-G-SIB banks will be less attractive than that of their G-SIB rivals. Duncan Martin, senior partner at the Boston Consulting Group and co-author of a report for the World Economic Forum on Rethinking Risk Management in Financial Services, says: "There is already an inverse relationship between size and the cost of wholesale funding and the creation of G-SIBs will simply reinforce it."
Where does that leave the non-G-SIBs? Oliver Judd, a credit analyst covering the banking sector at Aviva Investors, thinks clarity over who is and is not a G-SIB is important and will make investment decisions more straightforward. That, however, is not good news for non-G-SIBs. "Saying that these G-SIB banks are too big too fail is a danger for the market. Investors might well end up focusing on those 28 names and we could see a crowding out effect," says Judd.
Whether or not G-SIBs will get cheaper funding because of the moral hazard of an implicit state guarantee, or because they are carrying more capital, or because they are simply big and their debt is liquid, it is the likely outcome. The cost of greater equity will be offset, at least in part, by cheaper wholesale debt funding.
Equity vs debt
Changing the capital structure of banks will also distribute risk and reward differently among debt and equity holders. In the immediate aftermath of the financial crisis the interests of debt and equity holders were broadly aligned. The imperative was for banks to cut their risk exposure. Christoffer Mollenbach, managing director, head of financial institutions, debt capital markets at Lloyds Bank Corporate Markets can sense a change in mood.
"The interesting thing is that investors are divided into two camps depending on the asset class they are investing in," he says. "This presents a challenge for banks and is an evolution in the marketplace. Equity investors are unhappy with the pressure on NII. But debt investors like de-risking because it increases the certainty they will collect their coupons and get paid back at par."The debt of non-G-SIBs may be less attractive because they are not part of the too big to fail club. However, their equity might be more attractive. The higher equity buffers that the GSIBs will need to hold will mean many will have to raise capital. Mediobanca Securities estimates that European banks will need to raise 62bn of new equity capital to meet the G-SIB requirement, with Deutsche Bank, BNP Paribas, Crédit Agricole and Société Générale to the fore.
Though this is a drop in the ocean compared to the overall total of $2tr that the Institute for International Finance estimates will need to be raised by the banking industry worldwide to comply with Basel III, existing bank shareholders face the prospect of being diluted. Non-G-SIB banks, however, do not need to increase their core equity tier one capital beyond the 7% Basel III level unless this is deemed necessary by national regulators.
Shareholders of well capitalised non-G-SIB banks do not have to worry in the short term about dilution, which should make their shares more attractive. The less complex and less risky business models of the non-G-SIBs are other positives. They should also enjoy more flexibility over their capital mix.
The caveat though is that they are already well capitalised. When the European Banking Authority performed its stress tests in July it made it clear banks should be required to raise capital by any means, "including where necessary restrictions on dividends, deleveraging, issuance of fresh capital or conversion of lower quality instruments into core tier one capital." That raises the spectre of conversion into equity for debtholders and dilution and meagre dividends for equity investors.
When considering the pros and cons, the advantages of being outside the club may just be outweighed by the disadvantages. That is the view of Mollenbach at Lloyds. "I think a non-G-SIB is incentivised to increase its balance sheet size so it will look more like a G-SIB," he says.
Though there are only 28 G-SIBs, there are 73 banks that will be subject to substantially higher and continuous supervisory monitoring. On an annual basis the Financial Stability Board will score banks using its five criteria of size, interconnectedness, global (cross-jurisdictional) activity, complexity and the substitutability of trade.
There may be an incentive for banks which are just outside the G-SIB club to get just small enough that they will not have additional capital dragging down returns, but just big and complex enough to be too big to fail. Another advantage of being on the fringes of the club is that the other side of the coin of G-SIB status is that these banks must produce and keep updated resolution and recovery plans, so called living wills, which set out what will happen at the point of non-viability.
Boston Consulting Groups Martin says: "There are now two types of capital. Wholesale funding in the debt markets and hybrid capital, which can be thought of as going concern capital. The equity piece and G-SIB buffer is gone concern capital." However, though non-G-SIB banks will not have to produce living wills for the Financial Stability Board, it is highly likely they will for their national regulators. There are already regimes in place in the UK, US, Germany and Denmark and European Union proposals are expected soon.
How the G-SIBS and non-G-SIBs are treated by the rating agencies is likely to be a far more important driver of behaviour than living wills. At Moodys Investors Service, analyst Carlos Suarez Duarte says the greater clarity over how to define a G-SIB is positive from a ratings perspective because it removes uncertainty.
But he does not believe G-SIBs will receive preferential ratings treatment so long as the Financial Stability Board can complete its work on living wills and resolution mechanisms in tandem with defining which banks are in the club. Walter Pompliano, managing director, FI, DCM ratings advisory at Lloyds Bank Corporate Markets and a former employee of Standard & Poors, is not so sure.
"Banks that are not viewed as being systemically important are being downgraded. Who is in and outside of the club will be incredibly important for ratings," says Pompliano. Ratings will also drive the cost of funding so it seems this is yet another incentive to get bigger, despite the additional cost of carrying more equity capital.
Boston Consulting Groups Martin thinks those on the fringes of the G-SIB club will try to use their status more subtly. "Financial markets are like biological systems, they adapt and evolve. If the non-G-SIBs have a more expensive cost of capital they will find ways to shoehorn inappropriate assets on to the balance sheet and juice up returns. That is the nature of the beast."
That does not sound like a recipe for increasing financial stability. Simon Gleeson, a partner of Clifford Chance specialising in financial law and regulation is a little more hopeful. He thinks that the banks on the fringes of the G-SIB club will seek to look even more like a G-SIB bank in order to get similarly favourable treatment from rating agencies and from funding markets.
"Lets say you are bank number 29, 30 or 31. If I am the CEO and ask myself the question, what do I do in order to finance my business at no worse a rate than the smallest G-SIB? I think the answer is to hold even more capital than the G-SIB does because it has X amount of capital plus the benefit of the implicit state guarantee. Therefore, I would hold Y high quality capital, where Y is greater than X," says Gleeson.
If Gleeson is right and the non-GSIBs end up holding even more capital than their G-SIB rivals in order to maintain access to competitively priced wholesale funding, this may or not represent unintended consequences at work. The more capital that there is in the system, the more capacity there is to absorb losses, an outcome which may be attractive for regulators.
However, it is difficult to see how this outcome is compatible with non-G-SIB banks taking market share from their G-SIB rivals. Isaac Alonso, managing director and head of financial institutions capital origination at UniCredit, thinks non-GSIB banks should try to find positives, particularly when it comes to their capital structure.
"I simply dont agree with the people that say that non-G-SIBs should try to become G-SIBs. They should use their greater flexibility to their advantage. For example, regulators have said that high trigger contingent capital is not appropriate for G-SIBs but that leaves non-G-SIBs some leeway to at least have that discussion with their local regulators. If there is a funding penalty to being a non-G-SIB, I would look to educate investors and come up with a capital mix that would allow the bank to fund at attractive rates," says Alonso.
Avivas Judd says he would be prepared to look at hybrid capital, so long as the bank was already well capitalised. "There are some jurisdictions, potentially including the UK, where regulators may demand domestic buffers on top of the G-SIB buffers. In that instance we could see a role for hybrid capital and Cocos. Likewise, if a non-G-SIB entity ticked all the boxes in terms of other capital, hybrids could be attractive," says Judd.
For non-G-SIB banks that do want to get bigger, acquisition is the obvious route. The Financial Stability Board says that if "two banks merge and the resulting bank becomes a candidate for a different treatment within the G-SIB framework, this will be captured through the annual supervisory judgment process."
However, investors are likely to punish banks if acquisitions do not make strategic sense or are driven by the desire to enter the G-SIB club. "If bank management were making acquisitions, or increasing their balance sheet, or entering more cross-border or complex businesses in order to become a G-SIB that would be odd. We would certainly view it as very negative," says Avivas Judd.
Nevertheless, many expect that the banking sector will increasingly bifurcate. There will be the largest banks with access to wholesale funding markets. This will include the G-SIBs and the broader universe 73 banks that are under FSB supervision. These institutions will have coverage from the rating agencies and sell-side research. At the other end of the spectrum there will be small institutions reliant on deposits.
Banks that regard themselves as middle ranking but are outside the global 73 will probably have to get smaller. "Analysts have been saying for some time that mid-tier banks will need to get bigger or smaller," says Gleeson at Clifford Chance. "What the creation of the G-SIBs does is take away the choice. The new orthodoxy is that mid-market banks must get smaller."
There is also likely to emerge a third category of banks, not G-SIBs but national champions. They would still be able to access wholesale markets, though via a domestically oriented investor base. "In the UK I can think of a name like Nationwide," says Avivas Judd. "They are not even on the fringes of being a G-SIB but it is important in a UK context and would still be attractive for an investor like Aviva with a good understanding of their business and large sterling liabilities."
With markets in turmoil in August and 60,000 employees at just eight banks being sacked in a few short weeks, the notion that even bigger banking behemoths and a spurt in M&A will be two outcomes of the creation of the G-SIB club is fanciful at the moment. It was certainly not the intention of regulators. But some fear this will inevitably be the long term result as G-SIBs use their funding edge to drive up margins.
Boston Consulting Groups Martin says that those on the fringes of the club will look on enviously and make ill-judged acquisitions. "What will happen is nouveaux riche dumb money will enter investment banking and try to compete with the G-SIBs because they are making lots of money. Some will want to be G-SIBs, others will lobby hard not to be and will play games with their balance sheets to pretend they are not taking risk when they are. I doubt that the result is a safer system."Investors are more concerned about the present. "Debt funding may get cheaper over time, not because of greater capital or a safer financial system, but because these markets will hopefully become a bit more stable and normal," says Avivas Judd. "G-SIB or non-G-SIB is a bit of a moot point at the moment. In a more normal market environment it will become important. Right now our investment priority is safety first."