## Paul Tucker: Shadow banking - thoughts for a possible policy agenda
Speech by Mr Paul Tucker, Deputy Governor for Financial Stability at the Bank of England, at the European Commission High Level Conference, Brussels, 27 April 2012.
* * *
It is excellent that the EU Commission has published a consultative paper on shadow banking and is holding this conference today. The Commission's paper fits well with the approach that the G20 Financial Stability Board is taking. The issues here are very important since, as the international community reregulates the banking industry, more activity is almost bound to be booked outside banking.
We need at some point to move on to policy. I am therefore going to use today's occasion to put on the table some thoughts for a possible concrete policy agenda. I am doing so in the spirit of wanting to help nudge this work into its next stage. And I should be clear that the FSB, where our work is led by Adair Turner, have not yet articulated our collective preferred policy approach.
The definition of 'shadow banking' employed by the FSB and the EU Commission is, paraphrasing, credit intermediation, involving leverage and maturity transformation, that occurs outside or partly outside the banking system. 1 That is very close to the definition I used in a speech two years ago calling for work on shadow banking, except that in addition to leverage and maturity transformation I include 'monetary services'. 2
I want to make three broad points about this. First, it is very clear that 'shadow banking' is not the same as the non-bank financial sector. This is not a debate about the appropriate regulatory framework for the whole of finance. For example, the vast majority of hedge funds are not shadow banks, and don't trade in the credit markets or especially illiquid markets.
Second, non-bank intermediation of credit is not a bad thing in itself. Indeed, it can be a very good thing, helping to make financial services more efficient and effective and the system as a whole more resilient. We must remember that as we make policy.
But, as we know from this crisis and from previous ones, true shadow banking can weaken the system.
Regulatory arbitrage, which is always with us, can distort and disguise channels of intermediation.
That leads me to my third general point. Shadow banking comes in lots of shapes and colours. There are degrees to which any particular instance of shadow banking replicates banking. The liquidity offered by some shadow banks relies pretty well entirely, and more or less openly, on committed lines of credit from commercial banks. In these cases, the liquidity insurance offered by the shadow bank is 'derivative'; there is a real bank in the shadows. But for other shadow banks, liquidity services are offered without such back-up lines. In those cases, claims on the shadow bank have, in effect, become a monetary asset. Examples probably include money market mutual funds and an element of the prime brokerage services offered by securities dealers to levered funds.
1 See Section 1, Financial Stability Board, 'Shadow Banking: Scoping the Issues', April 2011; Section 3, European Commission , 'Green Paper on Shadow Banking', March 2011; and also Turner, 'Securitisation, shadow banking and the value of financial innovation', April 2012.
2 My definition was: instruments, structures, firms or markets which, alone or in combination, replicate, to a greater or lesser degree, the core features of commercial banks: monetary or liquidity services, maturity mismatch and leverage. See Tucker, P M W, 'Shadow Banking, Financing Markets and Financial Stability', January 2010.
With that brief bit of scene setting, I will quickly set out my current thinking on policy. The objective should not be to curb non-bank finance in general, but to recognise where a form of intermediation is banking in substance or in the risks it creates for the system as a whole.
## Shadow banks that are really part of banks
Many examples of shadow banking are sponsored by banks or are operated by them, or both. They are effectively part of their 'parent' bank. In the run up to the present crisis, prominent examples were Structured Investment Vehicles, ABCP conduits, and money funds. Many benefitted from financial support from their 'parent' during 2007-08. For such situations, the following could usefully apply:
- Shadow banking vehicles or funds that are sponsored or operated by banks should be consolidated on to bank balance sheets
Consolidation might require changes in accounting rules. That could take time. These vehicles and funds should nevertheless be treated as consolidated in the application of Basel 3 regulatory capital requirements etc. If necessary, Pillar 2 should be used to achieve that. (The Basel Supervisors Committee will, I hope, consider this in its contribution to the FSB work).
## Banks' provision of committed lines of credit to independent shadow banks
Even where a shadow bank is not de facto or de jure part of a banking group, many are fundamentally dependent on banks - through committed lines of credit. Anyone running a maturity mismatch is exposed to liquidity risk - liabilities being called before assets fall due or before they can be sold in an orderly way. Banks can provide insurance against such liquidity risk because their deposit liabilities are money; they can lend simply by expanding the two sides of their balance sheet simultaneously, creating (broad) money. But, for the system as a whole - ie taking a macroprudential perspective - providing committed lines to shadow banks is riskier than providing such lines to non-bank businesses. Shadow banks are liable to call on their lines just when the banking system is coming under liquidity pressure itself. That should be reflected in regulatory policies on banks' liquidity exposures. I suggest that:
- the draw-down rate assumed in the Basel 3 Liquidity Coverage Ratio should be higher for committed lines to financial companies than for lines to non-financial companies. That is, banks should hold more liquid assets against such exposures.
Without a policy of that kind, there will be a strong incentive for maturity-transformation to be undertaken off bank balance sheets but with an umbilical cord back to the banking system. It should apply not only to vehicles used for securitisation but more generally.
## Money funds
Money funds do not use committed lines of credit from banks. Claims on money funds have, in effect, become monetary assets in the hands of savers. In parts of the world, especially the US, they are treated like current accounts. Given the restrictions on their asset holdings, they resemble narrow banks, in mutual-fund clothing. But for a normal mutual fund, as an open-ended investment vehicle, the value of investments in it fluctuates with the value of the vehicle's asset portfolio. By contrast, most money funds hold themselves out as offering par under any circumstances; when they 'break the buck', they must unwind. Their investors run at that prospect; and so the funds themselves are flighty investors. Compared to most types of shadow banking, money funds do not borrow - in the usual sense. But by promising par, they are in effect incurring debt-like obligations. And they can be exposed to leverage. At least in the run up to the crisis, some invested in levered paper, some of it in what amounted
to Russian Doll shadow banking - a money fund buys short-term ABCP backed by CDOs, etc.
What I suggest here is that:
- Money Market Funds should be required to choose between being
- -Variable Net Asset Value (NAV) funds or Constant NAV funds
- -Any remaining CNAV funds should be subject to capital requirements of some kind
- -All should be subject to 'gates' or other measures that can be used to delay withdrawals, to make runs less likely.
That package would not completely prevent runs; regular mutual funds can suffer runs. But it would make them somewhat less brittle.
Europe should want a global standard or at least a globally consistent approach to money funds. Much the largest money-fund industry in the world is based in the US. But it is not confined to domestic, US intermediation. It is internationally active - lending to banks, corporates and sovereigns around the world. That underlines the importance of the outcome of the deliberations of the SEC and of the US Financial Stability Oversight Counsel in this area. We have to hope that the US authorities will take action - making not only their own domestic financial system safer, but the global system safer.
If, against our hopes and expectations, that proves not to be the case, authorities in Europe and elsewhere will need to think through what if any measures we could sensibly take to make our part of the global financial system more resilient to the faultline that the money fund industry currently represents. One possibility would be for
- bank supervisors to limit the extent to which banks could fund themselves short-term from US money funds and from other fragile/flighty sources, including CNAV money funds domiciled elsewhere.
I should stress that such a policy need not be targeted at US funds per se. It is also a question of the prudence of the banks that borrow. So it could be cast in terms of the liability structure and sources of funding of the banking system. The Basel 3 Net Stable Funding Ratio may help a bit with this.
And I should add that on some fronts - notably transparency - Europe should aim to catch up with the progress the US has already made.
## Securities dealers, finance companies
Some shadow banks are 'businesses', not funds or vehicles. These non-bank financial intermediaries finance themselves externally in the markets. Some such financing is short-term and so effectively runnable, as events showed. In those cases:
- If they are financed materially by short-term debt, they should be subject to bank-type regulation and supervision of the resilience of their balance sheets.
That would apply to some kinds of finance company. It might apply to securities dealers too.
In the EU, many dealers are 'investment firms' and so are already subject to the same Directives as banks on capital adequacy. That is not true of finance companies in many jurisdictions.
Perhaps there should be a size threshold for applying bank-style capital and liquidity regulation to such non-bank banks.
## Securities dealers and the 'rehypothecation' of client assets
That last set of points was about firms acting as principals. Major issues arise from the use of client assets - money and securities - to finance the principal risk-taking of some financial firms. This has become apparent from the Lehman and MFG bankruptcies. Quietly in the background, something extraordinary has happened over the past decade or so. Let me explain.
You place £X of your securities portfolio with your broker/dealer in safekeeping (as a subcustodian in a way). You borrow a much smaller amount of money from your broker and, reasonably enough, they take some of your securities as collateral; in effect, they have financed your holding of that part of your securities portfolio. An observer might think the rest of your portfolio would sit in a segregated client account. Not a bit of it. Unless you have been very careful, the broker/dealer has repoed out your securities for cash, and used the proceeds to finance its own businesses. To make this simpler, imagine you have placed surplus cash with your broker-dealer. As likely as not, it is being used to finance the dealer's business.
Whether we are talking about your cash or your securities, if you can call them on demand and they are being used to finance your broker's business, that used to be called ..... banking.
I suggest the following in this area:
- Only banks should be able to use client moneys and unencumbered assets to finance their own business to a material extent; and that should be a clear principal relationship . Legal form should come into line with economic substance.
- For non-banks, any client moneys and unencumbered assets should be segregated and should not be used to finance the business to a material extent. It should, however, remain permissible for non-banks to lend to such clients on a collateralised basis to finance their holdings of securities ('margin lending').
This is an area where regimes vary enormously around the world or are non-existent. For example, the US domestic regime is stricter on this than Europe. We need a global discussion.
## Securities lending, repo and collateralised-financing markets
So far I have discussed funds or vehicles or firms that are in economic substance shadow banks. We also need to think about the markets that firms and vehicles can use to develop a de facto banking business. Perhaps the key such markets are securities lending and repo or, more generally, collateralised borrowing.
Before proceeding I must stress that these markets are vital to efficient capital markets.
The crucial economic function of market-making entails intermediaries going short, and so needing to cover their position by borrowing securities. And repo has enriched the money markets.
But it is also true that anyone holding a securities portfolio can build themselves a shadow bank using the securities lending and repo markets. One simply lends out the securities at call for cash, and then one employs that cash by making loans or buying credit-assets with a longer maturity. This is leverage and maturity mismatch.
That example assumes the shadow-banking business is built by the investor who holds the securities outright. But, of course, nothing is so simple. The first lender of securities might lend against securities collateral and do no more; that is relatively common in European markets. But the entity that has borrowed those securities could themselves repo out the borrowed securities for cash, and employ the cash in a lending or credit-asset business. And
BIS central bankers' speeches
so on - ie at any point in the chain of securities lending, an intermediary can build themselves a shadow banking business. Such chains can prove fragile for all sorts of reasons - because many securities are at call, and because many secured lenders try to realise their collateral instantly upon default.
It is also worth mentioning that some asset managers have no - zero - appetite to hold the underlying paper outright in the event of their counterparty defaulting, either because the assets are not covered by investment mandates or the fund managers do not know how to manage them, etc.
What I suggest on this front is:
- There should be greater market transparency, perhaps ideally via a Trade Repository with open access to aggregate data, so that the world can see what is happening in these very important but opaque financing markets. (That would be helpful for market participants themselves).
- Financial firms and funds should not be able to lend against securities that they are not permitted or proficient enough to hold outright.
- Non-bank financial firms should be regulated in how they employ cash collateral.
- The authorities should be able to step in and set minimum haircut or margin levels for the collateralised financing markets (or segments of them). (That would need to be pursued at international level. 3 It might be linked to central bank haircuts).
Some of those points are aired in the FSB's report on securities lending and repos published today 4 and will, I hope, be explored intensively in the coming months.
I will expand slightly on the last point in my list. The amount of leverage available in the system depends, of course, to a significant degree on the haircuts applied in chains of secured financing transactions; leverage is the reciprocal of the haircuts. The resilience of the system is therefore impaired when a bout of exuberance drives down haircuts. The authorities need to be ready to place constraints on that - through minima that either apply continuously or are specially introduced when conditions are overheating.
There is already an official exercise underway to frame the determination of minimum margin levels in the parts of the derivatives market that will not be centrally cleared. At some point down the road, this will need to be brought into a coherent framework with any measures on the collateralised-financing market. Doing so would require deciding whether to apply the same or different approaches to the inter-dealer market and dealer-customer transactions.
## Conclusion: innovation, evolution and surveillance
Those are just a few thoughts for debate, based on some current manifestations of shadow banking. I have not even covered the existing waterfront. For example, we need policies on those levered credit funds that offer near-instant withdrawal rights. In the EU distinctions need to be made within the Alternative Investment Fund Managers Directive and UCITs regime, which allows risky structures to be branded as retail products.
But it would be foolhardy to imagine that we can frame policies today that will stand the test of time. The financial system will evolve, and we need to permit innovation. A policy framework on shadow banking therefore needs to be adaptive. And it mustn't try to shut
3 See Committee on the Global Financial System, 'The role of margin requirements and haircuts in procyclicality', March 2010.
4 See 'Securities lending and Repo: Market Overview and Financial Stability Issues', and interim report of an FSB group chaired by my colleague David Rule.
everything down. As I said at the outset, non-bank finance is not intrinsically a bad thing. We will need effective surveillance of what is going on and what concentrations of risk are emerging - through outfits like the FSB's committee on vulnerabilities and, in the EU, through the European Systemic Risk Board. And we will have to make discerning policy judgments that are explained and consulted upon. That is exactly what the EU, alongside the FSB, is now embarked on.