## Thomas Jordan: Macroprudential instruments and balance sheet risks
Introductory remarks by Mr Thomas Jordan, Vice-Chairman of the Governing Board of the Swiss National Bank, at the end-of-year media news conference, Zurich, 16 December 2010.
* * *
In my following remarks, I will address two issues in depth. I will start by detailing the set of instruments which the Swiss National Bank (SNB) should have at its disposal for the prevention of financial crises. I will then take a look at the risks on our balance sheet.
## Macroprudential instruments at the SNB
More than three years have passed since the outbreak of the international financial crisis. The crisis made two things very clear. First, the capital and liquidity buffers in the financial system were far too low. Second, systemic risks had been grossly underestimated prior to the crisis or had not even been identified. Almost nobody suspected that the risks were so closely interlinked, nor that problems in a small segment of the US mortgage market could rock the global financial system and the world economy. In order to make the financial system more resilient, one thing that needs to be done is to strengthen the capital and liquidity buffers. Significant progress has already been made in this regard with the approval of Basel III and the new 'too big to fail' (TBTF) regulations. These measures should help ease the problems in connection with the TBTF issue significantly. But there will be other crises in the future, too. That is why it is equally important to identify the build-up of risks in the financial system at an early stage and to take countercyclical measures to mitigate them. It is in this area - referred to as macroprudential policy - where there are still major deficiencies.
The National Bank has the statutory mandate to contribute to the stability of the financial system. In its function as lender of last resort, the SNB plays a key role in managing financial crises. It can provide emergency liquidity assistance to a bank that runs into difficulties, thus preventing the negative impact the bank's default could have on both the banking sector and the economy. This remedial measure proved highly effective during the crisis. However, emergency liquidity assistance should only be provided in exceptional cases, as it creates false incentives for banks and entails considerable financial risks for the SNB. In order for the National Bank to fulfil its mandate, it is imperative that it also be able to take more preventative action in the future. This is the only way the SNB can guard effectively against future crises.
Preventative action is only possible if the National Bank has efficient and effective instruments at its disposal. However, there is currently a glaring deficiency in this area. Up until now, the only precautionary measure the SNB could employ was to issue a warning. Yet experience has shown that warnings alone are not enough. Before the crisis, the National Bank had repeatedly drawn attention to the high leverage of the big banks and warned that their ability to absorb losses was not adequate. These caveats did not result in any voluntary change in behaviour. For this reason, it is essential that our set of preventative instruments for financial stability be improved in three areas; namely, improved access to information, closer involvement in regulatory procedures with regard to financial stability issues, and direct macroprudential decision-making powers. I would now like to take a closer look at these.
First, it is crucial that we have all the necessary information quickly to hand in order to identify the build-up of risks in the financial system at an early stage. While the SNB's statutory authority to compile statistical data is both necessary and helpful, the latest crisis clearly demonstrated that the National Bank must be able to request additional information. With this I mean, for example, quantitative, non-standardised data on banks' risk exposure.
Unlike FINMA, Switzerland's financial market supervisory authority, the SNB's stance is primarily a macroeconomic or systemic one. This is why we must be able to act independently in certain situations when gathering information. At the same time, we also want to keep our access to information to an absolute minimum and avoid a duplication of activities with FINMA.
Second, clarification of the SNB's participation rights and obligations is needed in the area of regulation essential for financial stability. In order to give appropriate consideration to financial stability aspects, we feel that the SNB should be involved in regulatory procedures at an early stage. In other words, in the conceptual phase. The authorities responsible for issuing financial market regulations should, therefore, be obliged to consult with the SNB early on in the process.
Thirdly, the National Bank must have clearly defined decision-making powers with regard to regulations that either bear direct relation to monetary policy or to emergency liquidity assistance. I would like to take two current regulatory proposals to explain more clearly what I mean: the countercyclical capital buffer and the implementation of the TBTF regulations.
The countercyclical capital buffer represents a capital requirement that can be adjusted over time. This buffer will be introduced within the framework of Basel III and has the important tasks of protecting the banking sector from the risks posed by excessive lending and of curbing the banking sector's procyclical behaviour. In the event of excessive growth in lending, the capital requirement can be increased; then when growth in lending is weak, it can be lowered again or even reduced to zero. The countercyclical capital buffer thus has a direct impact on banks' lending activities and can, as a result, strengthen or weaken the effect of the SNB's monetary policy significantly. Therefore, whoever determines the level of this buffer also influences the effect of monetary policy. It is for this reason that - in our opinion - the responsibility for the countercyclical capital buffer for domestic lending must lie with the SNB. The National Bank must be able to decide when the buffer is activated and how high it should be. It goes without saying that it would do so within legally defined limits. The same should also apply to the activation of loan-to-value ratios and affordability criteria for mortgage loans. All of these instruments are closely linked to monetary policy.
Within the context of the implementation of the TBTF regulations, the systemically important banks will have to be determined. This is because the more rigorous capital and liquidity requirements are only intended for these banks. And because emergency liquidity assistance is also reserved for systemically important banks, the decision as to the systemic importance of a bank has a direct impact on the SNB in its function as lender of last resort. These decision-making powers, too, must therefore be placed in the National Bank's hands.
This proposed solution should give the SNB the formal responsibilities it needs in order to fulfil its mandate in the area of financial stability in a preventative manner. In many countries, a transfer of banking supervision to the central bank is currently the subject of intensive discussion. In contrast, the National Bank is by no means calling into question the separation of banking supervision and central bank activities with the solution presented here. This separation has worked for Switzerland. The aim is not to infringe on FINMA's jurisdiction. Instead, FINMA and the SNB need to clearly define how the new responsibilities and tasks arising from the urgent need for more rigorous analysis and regulation of systemic risk should be divided between them. At this point, I also want to stress that this macroprudential task does not call into question the SNB's primary mandate, which is the maintenance of price stability. Instead, it supports the long-term fulfilment of this mandate. The SNB has also expressed its opinion - along the lines of this proposal - on the recommendations made by the business audit commissions of the National Council and the Council of States in their report on the financial crisis.
BIS Review 173/2010
## SNB balance sheet risks
In my following remarks, I would like to look at the SNB's balance sheet risks. As the Chairman of the Governing Board, Philipp Hildebrand, has already mentioned, it was primarily the large-scale purchases of foreign currency that contributed to the expansion of our balance sheet to its current level of approximately CHF 280 billion. Of this amount, CHF 260 billion is made up of currency reserves, which comprise not only foreign exchange, but also gold and the IMF's special drawing rights. If we add to that the SNB's loan to the stabilisation fund, over 95% of our assets are denominated in foreign currency.
Not only is the balance sheet longer as a result, it also carries more risk. Without doubt, the exchange rate risk is now our most significant risk. While fluctuations in annual earnings running into the billions have also occurred in the past, they are likely to be even more pronounced in the future owing to the greater size of the balance sheet. In this context, I would like to point out two factors in particular.
First, experience has shown that losses on foreign currency positions decline over time. On the one hand, strong currency fluctuations are often followed by corrections, since markets tend to overreact, while on the other, interest and dividend payments help to reduce losses on foreign currency investments.
Second, the SNB has, for a long time now, placed a great deal of value on a robust balance sheet. Established mechanisms for the steady build-up of provisions and smoothing of profit distribution have helped build up an equity capital buffer over the years, and this makes it possible for us to carry losses. Owing largely to this foresighted approach, the level of equity capital on our balance sheet is, by international standards, still above average, in spite of its current size.
A gratifying reduction in the balance sheet risk was achieved in 2010 for the stabilisation fund. Given that the market situation improved considerably this year, holdings of some USD 3 billion were sold and a number of derivative positions were closed. In addition, interest payments and repayments on investments amounted to approximately USD 3 billion. Overall, the SNB's loan to the StabFund at the end of November 2010 was down to just roughly USD 13 billion, compared to almost USD 20 billion at the end of 2009. At the same time, the overall risk for the SNB decreased from USD 24 billion to USD 16 billion. This USD 16 billion corresponds to the maximum possible loan. This upper limit exceeds the current loan, since part of the StabFund's portfolio does not require any financing at present. Owing to the continued favourable market situation in the fourth quarter, the value of the fund's assets is now considerably higher than the value of the loan. The risk of the SNB suffering a loss upon liquidating the fund has thus decreased further.
As long as we have to maintain high levels of foreign exchange reserves for monetary policy reasons, the currency risks in our balance sheet will not be easy to reduce. We have, however, taken two measures in an effort to strengthen our balance sheet further. First, we decided back in 2009 to double the annual allocation to provisions over the course of five years. Second, in recent months, we have reduced risk concentrations in currency holdings and borrower categories - which arose from the interventions - through diversification. My colleague, Jean-Pierre Danthine, will discuss this in greater depth.
As previously mentioned, the higher level of risk will be evidenced by greater fluctuations in the SNB's annual earnings. It is too early to say whether this will eventually affect the distribution of profit to the Confederation and the cantons. There is no need for action at this stage. The profit distribution agreement between the Federal Department of Finance and the SNB sets out clearly defined threshold values for a review. The current agreement - which runs until the end of 2017 - will be reviewed at the latest with regard to the distribution for the 2013 financial year. Should the distribution reserve, which has been built up over the last few years, be depleted before then, the agreement calls for an immediate review.
It is currently too early to say anything definite about the outcome of the review. Seen from the current perspective, it would come as no surprise, however, if the annual distribution had to be reduced somewhat. We have repeatedly emphasised that an annual distribution of CHF 2.5 billion cannot be guaranteed in the long term.
BIS Review 173/2010