Lars Rohde: Denmark's fixed exchange rate policy
Speech by Mr Lars Rohde, Governor of the National Bank of Denmark, at the annual meeting of the Danish Bankers Association, Copenhagen, 7 December 2015.
The views expressed in this speech are those of the speaker and not the view of the BIS.
Thank you for inviting me to speak.
I have the following key messages today:
In view of the special circumstances relating to the krone at the beginning of the year, and the ECB's decision last Thursday to reduce its deposit rate further, I would like to make clear that nothing has changed as regards Danmarks Nationalbank's willingness and ability to defend Denmark's fixed exchange rate policy.
My second message is that a prolonged period of low interest rates may lead to high asset prices. But subsequently the assets may fall considerably in value, which could result in losses, also for lenders. Furthermore, there is a risk that the low interest rates will impede productivity development in Denmark as they will buoy up unprofitable firms that would otherwise close.
Finally, I would like to emphasise that under the new EU recovery and resolution regime for banks, Danmarks Nationalbank remains the lender of last resort. But the banks' liquidity management should not be based on access to emergency liquidity assistance from Danmarks Nationalbank if they come under pressure.
First, let me talk about the economic situation.
While growth in the emerging market economies has shifted to a lower gear, the advanced economies are still picking up steam. The USA is in a robust upswing, with rising employment and falling unemployment. In the euro area, virtually all member states are growing at a rate above their potential, so that they are approaching a normal level of capacity utilisation.
In Denmark, growth will be a little lower than previously forecast this year, but the Danish economy remains on track and spare resources are likely to become more scarce in the next couple of years. This is seen most clearly in the labour market, where employment is expected to continue to rise, while unemployment is declining.
As regards monetary policy, the respective authorities in the USA and the euro area are currently pursuing opposite paths. While the market expects the Federal Reserve to respond to a strong labour market by increasing interest rates soon, the ECB last Thursday reduced its deposit rate by 10 basis points to -30 basis points. In addition, the existing asset purchase programme has been extended. Easing of monetary policy in the euro area is aimed at ensuring that the ECB achieves its target of inflation below, but close to, 2 per cent.
In the light of the ECB's monetary policy steps, Danmarks Nationalbank has chosen to keep its monetary policy interest rates unchanged. From April until and including November, our intervention sales of foreign exchange in the market totalled kr. 230 billion, and therefore it is appropriate to allow the interest rate spread between Denmark and the euro area to narrow a little.
We have plenty of patience if it should once again become necessary to intervene to prevent the krone from strengthening. In that situation - as at the beginning of the year - there will be no upper limit to the amount of kroner that Danmarks Nationalbank can make available. We have no target for the size of the foreign exchange reserve. As long as short-term interest rates are lower in Denmark than in the euro area, the reserve even yields a positive return. Furthermore, we can reduce Danmarks Nationalbank's interest rates if necessary.
Since March this year, the foreign exchange reserve has shrunk substantially as a result of intervention in the market, and by the end of November it was kr. 484 billion. That is not much higher than the level before the large inflows of capital in January and February. The capital inflows prompted Danmarks Nationalbank to intervene in the foreign exchange market for very large amounts, and our interest rates were also reduced considerably.
The gradual reduction of the foreign exchange reserve reflects that investors have sold kroner since the large inflows of capital at the beginning of the year. But they are not selling them as fast as they have done. All in all, the foreign investors who bought kroner at the beginning of the year have sold them again. Domestic investors have also reduced their holdings of kroner, but at a slower pace, and recently their holdings have increased again. Against that background, the Ministry of Finance, upon the recommendation of Danmarks Nationalbank, has resumed issuance of government bonds. So far, five auctions have been held, and demand has generally been high.
10 months have now passed since we reduced the rate of interest on certificates of deposit to -0.75 per cent. At that time we were moving into unknown territory: never before had Danmarks Nationalbank's interest rates been so low, and internationally only Switzerland had introduced a corresponding level a short while earlier. The low interest rates have had the desired effect. The krone has remained stable against the euro and there have not been any signs of cash hoarding.
Historically, monetary policy interest rates have been a little higher in Denmark than in the euro area. But right now the interest rate spread to the euro area is negative. This means that investors are willing to hold financial assets in Danish kroner even if the return is a little lower than for euro investments. The long-term interest rate spread between Denmark and Germany is positive, however. Like Denmark, Germany has a robust economy and current account surpluses. Furthermore, interest rates in Germany are to a large extent affected by the ECB's asset purchase programme.
The reduced interest rate spread to the euro area may be attributable to Denmark's build-up of considerable net foreign assets in recent years. The increased wealth gives rise to various payment flows that tend to strengthen the krone, thereby reducing the interest rate spread.
The accumulation of Danish net foreign assets is not a passing phenomenon. Consequently, there is reason to believe that the interest rate spread to the euro area will, in the long term, be lower than previously. Exactly how much lower it will be is difficult to say. Recent years have been characterised by extraordinary events such as the sovereign debt crisis in a number of euro area member states and, most recently, quantitative easing by the European Central Bank, ECB. So these years cannot be seen as an indicator of the situation in the longer term. All the same, it seems likely that Danmarks Nationalbank's interest rates will be very close to those of the ECB in future, and at times also a little lower.
Shifts in the willingness of domestic investors to hold unhedged positions in euro may lead to fluctuations in the size of Danmarks Nationalbank's foreign exchange reserve. It can become considerably larger than it has previously been. We have seen a clear example of that this year. If Danish investors are not willing to take on the foreign exchange risk linked to larger net foreign assets, it will fall upon Danmarks Nationalbank, and the interest rate spread becomes low or negative. This is part of the fixed exchange rate policy, which has been the foundation for a stable economy for more than three decades.
I will now turn my attention to credit quality. During the crisis we learnt, once again, that the value of the collateral pledged for loans may decrease considerably. This was the case for e.g. agricultural land, commercial properties and not least housing. In some cases the price dropped so much that the value of the collateral fell below the loan amount. As a result, a loan which looked safe at the time when it was granted might later result in large losses if the borrower defaulted and the collateral had to be realised. This led to losses for credit institutions, especially for the banks, as their loans are subordinated.
We should not forget that. At present this probably applies mainly to housing loans in the Copenhagen area. Here, the rate of price increase has been unsustainably high for some time, although quarterly price increases for owner-occupied flats have slowed down since the spring. The higher prices should be seen in the context of low interest rates and a rise in the population of the capital. One reason why this has so strong an impact on prices is the absence of automatic stabilisers in the form of countercyclical housing taxes. With the freeze on land tax in 2016, which will push up prices in the cities in particular, fluctuations will be even larger. That will increase the risk of the value of the home subsequently falling below the loan amount.
When the housing bubble burst, we saw that prices in Copenhagen can dive very sharply. All the same, the credit institutions' losses on housing loans were modest. This was mainly because of the low interest rates, which helped ensure that most homeowners were still able to service their loans. However, we cannot assume that interest rates will also be low the next time house prices fall.
The fact that interest rates may go up and house prices down should be a natural element in the assessment of potential loans. These are sound principles, which are also reflected in the new guidelines from the Danish Financial Supervisory Authority on "Prudent credit assessment when granting housing loans in growth areas".
The currently very low level of interest rates makes it cheaper for firms to service their debt. And the low level of interest rates is, so to speak, acting as a life-support mechanism for some of them. If interest rates rise, and the firms have variable rate loans, it may soon turn out that they are unable to survive.
The clearest example of this is the agricultural sector. In 2014, nearly 25 per cent of all farms were operating at a loss, even though they were being buoyed up strongly by the very low interest rates. If the rate of interest had been 2.5 percentage points higher in 2014, nearly half of all farms would have been in the red. That is, twice as many as was actually the case due to the very low interest rates.
An interest rate increase of 2.5 percentage points is not a lot. If the variable interest rate on mortgage loans rises by 2.5 percentage points, it will more or less match the current rate of interest on fixed rate mortgage loans. This indicates that close to one in two farmers cannot afford to remortgage into a fixed rate loan, which would otherwise provide protection against future increases in short-term interest rates. This emphasises how vulnerable the agricultural sector is - and that many farmers have only been saved by the very low interest rates.
The implications of the very low level of interest rates should be taken into account when a bank assesses how to deal with a defaulted loan. This applies to all loans - not just agricultural loans. In some cases it may be better to shut down the exposure rather than waiting and hoping that the firm will, despite tough odds, survive and be able to repay all its loans, including interest. Especially if the bank must finance a deficit to keep the firm afloat. That is bad for the bank, which risks eventually suffering a larger loss than it would otherwise have done. And it is bad for the economy because it impedes productivity development. Labour and capital are tied up in firms that are not viable, which means that they cannot be employed where they are most productive.
I will now address the issue of cyber risk; a topic that is receiving considerable attention both in Denmark and abroad. In an increasingly digitised world, society is dependent on IT systems functioning and data being protected. That also applies to the financial sector. It is well known that future problems may be of a type not previously seen, and the next challenge to financial stability could well be a cyber attack. For Danmarks Nationalbank, increased focus on cyber risk is a priority. I have put cyber risk on the agenda for the meeting of the Systemic Risk Council later this month.
Another topic currently receiving much attention is market liquidity. The focus is mainly on declining and less resiliant liquidity in the bond markets. High liquidity supports an efficient financial system. Danmarks Nationalbank regularly assesses whether changes in liquidity may have negative implications for financial stability and keeps up a dialogue with market participants. As we also do in other areas, we appreciate our close cooperation with the Danish Bankers Association and its members in this respect.
Market liquidity cannot be measured directly. Therefore, an assessment of liquidity must be based on a number of indicators. In the market for mortgage bonds, traditional liquidity measures do not show any clear signs of reduced liquidity. All the same, liquidity has been more volatile since late 2014, which indicates that it has become less resiliant. Hence, there is a risk that smaller shocks than previously may cause liquidity in the bond market to evaporate.
Market participants point out that market makers have become less willing to absorb imbalances between bond supply and demand. This should be seen in the context of lower risk appetite and increased capital and liquidity regulation since the financial crisis, which could mean that the banks reduce their market making in the bond market. Over the last year, Danish banks have reduced their portfolios of mortgage bonds for market making, which supports this view.
Liquidity in Danish government bonds fell in 2015. The same pattern can be seen in other European government debt markets, but Denmark was particularly severely affected due to the suspension of issuance. Since bond issuance was resumed in October, demand has generally been high, and liquidity has improved. Trading volumes are back at the level seen before the suspension of issuance. Price quotation has also become sharper, but there is still room for improvement.
Finally, I would like to mention that we - like the rest of the EU - now have a new recovery and resolution regime, the BRRD, which lays down uniform rules for handling distressed banks. The rules must be fully implemented by 1 January 2016. The fundamental principle is that the owners and creditors should bear any losses, not the government. This is a sound principle that is to eliminate the implied government guarantee for the largest banks.
Under the new set of rules, all banks are required to prepare recovery plans. These plans must describe how the bank can respond if it encounters difficulties. The bank should, to a far higher degree than previously, be able to take the necessary steps to restore its liquidity situation itself. A bank must never count on emergency liquidity assistance from Danmarks Nationalbank.
Danmarks Nationalbank is still the lender of last resort, as we have always been. This applies in connection with liquidity problems in the sector in general, but also if a single solvent bank has an urgent need for emergency liquidity assistance - ELA. The role of lender of last resort is an element of Danmarks Nationalbank's task of contributing to financial stability. But the banks' liquidity management should obviously not be based on access to emergency liquidity assistance from Danmarks Nationalbank if they come under pressure.
If a bank requests ELA, Danmarks Nationalbank as lender will assess whether the bank is solvent and creditworthy and thus able to repay the loan when it matures. Danmarks Nationalbank will include the Danish Financial Supervisory Authority's assessment of the bank's solvency and liquidity as a natural element in this connection. Furthermore, Danmarks Nationalbank will require adequate collateral for the amount granted.
No banks are precluded from ELA in advance. But besides considering solvency, creditworthiness and the ability to pledge collateral, Danmarks Nationalbank will assess whether financial stability considerations warrant ELA. This assessment will be based on the specific circumstances in the relevant situation. Under the new recovery and resolution regime, the Financial Stability Company, as the resolution authority, is given a number of resolution tools to prevent a resolution from causing financial instability. This means that there is a genuine alternative to granting ELA. We therefore expect the new recovery and resolution regime to limit the number of situations where ELA is relevant.
Let me conclude by reiterating my three key messages:
Neither the unusual circumstances relating to the krone at the beginning of the year nor the ECB's decision last Thursday to reduce its deposit rate further affects Danmarks Nationalbank's willingness and ability to defend Denmark's fixed exchange rate policy.
My second message is that the banks should be aware that a long period of low interest rates may entail credit quality risks, e.g. via fluctuations in the value of the collateral. The low interest rates may also impede productivity as they will buoy up unprofitable firms that would otherwise close.
My final message is that Danmarks Nationalbank is the lender of last resort, also under the new EU recovery and resolution regime for banks. But let me emphasise that the banks' liquidity management must not be based on access to emergency liquidity assistance from Danmarks Nationalbank.
Thank you.