Cecilia Skingsley: A year of negative interest rates. Where do we stand now?

Speech by Ms Cecilia Skingsley, Deputy Governor of the Sveriges Riksbank, at Danske Bank, Stockholm, 7 April 2016.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Thank you to Christina Nyman, Ulf Söderström, Charlotta Edler, Marianne Sterner, Emil Brodin and Anders Vredin for useful comments and assistance.

Thank you for inviting me here today. I intend to talk about three issues:

Is it possible to conduct a national monetary policy in a complicated international environment? How has Swedish monetary policy been conducted in recent years? What considerations should govern our choice of future path?

Is it possible to conduct a national monetary policy in a complicated international environment?

I believe that most people will recognise the description of growth and the recovery following the global financial crisis as being a disappointment in most countries. In many cases the setbacks have been severe and prolonged, for instance, in southern Europe, while in other cases growth has been positive but low.

Many central banks, wishing to live up to their price stability commitments, have implemented more far-reaching and long-term measures to stimulate monetary policy than have previously been imagined as possible.

So how did we get to where we are now, with such low interest rates?

In the long run, it is structural factors such as the conditions for growth and the willingness to save that determine the level of the real interest rate. As a small, open economy, Sweden must in principle accept the international real interest rate as a given. Monetary policy is not able to affect the real interest rate to any great extent in the longer run, but it affects inflation and inflation expectations.

In recent years, the IMF and other forecasters have revised down their forecasts for GDP on several occasions. If one looks at real interest rates, these began to fall even before the financial crisis and real interest rates have also fallen during periods of higher growth. Opinions differ with regard to the causes of this slowdown. The more pessimistic line talk about secular stagnation, where savings have increased and willingness to invest has declined as a result, for instance, of the composition of the population, of growing income gaps and weak technological innovations. This has in turn pushed down the interest rate level compatible with normal resource utilisation.

The other more optimistic line say that remaining effects from the global financial crisis, such as deleveraging and increased political uncertainty, explain the slowdown but that these effects are transitory.¹

Lower real interest rates and inflation

But it is not merely real interest rates that have fallen. In recent years, inflation has also shown a downward trend. This trend began later than the decline in real interest rates, but Swedish inflation began to fall slightly before that in other western countries such as the United States and the euro area. The most recent downturn in inflation in the euro area can to some extent be explained by the fall in energy prices. However, underlying inflation has also fallen in the wake of both the banking crisis and the euro crisis.

It is necessary to take into account the decline in both real interest rates and inflation in the monetary policy deliberations. The fundamental reasoning regarding the stabilisation ability of monetary policy with a floating exchange rate is based on the assessment of a normal nominal interest rate, simply described as an interval for a nominal interest rate that neither speeds up nor slows down the economy. To estimate how high the normal nominal interest rate should be, we look at the real interest rate and add to this the inflation target. If the real interest rate is 2 per cent and the inflation target is 2 per cent, then the nominal interest rate should be around 4 per cent. If the real interest rate rises to 3 per cent, and the inflation target remains unchanged at 2 per cent, the nominal interest rate will be 5 per cent, and so on.

But as I have just shown, developments are instead showing a downward trend. Lower real interest rates and lower inflation and inflation expectations are pushing the nominal interest rate downwards. This means that the central bank has to take stronger action for monetary policy to have a stimulating effect.

Some important questions are raised here: What are the fall in real interest rates and the downturn in inflation due to? Are these temporary or permanent changes? Are the central banks over interpreting their price stability task or have they not done sufficient to cure the low inflation? Some say that all things bad are mostly the fault of the central banks. I believe that part of the intensive debate on monetary policy conducted both here in Sweden and abroad is due to us economists tending to have different answers to these questions. These answers in turn affect the opinions on what is the best policy in both the long and the short term.

How has Swedish monetary policy been conducted in recent years?

Swedish monetary policy strategy entails stabilising inflation around the target and striving to stabilise production and employment around paths that are sustainable in the long term.

If we look back two years, to the period before the April 2014 meeting, CPIF inflation had been around 1 per cent for some time, which is too low given that the Riksbank wishes to uphold confidence in the inflation target of 2 per cent as an anchor for price-setting and wage-formation. Another observation from April 2014 was that the more long-run inflation expectations had fallen below the 2 per cent level. It became increasingly clear over the course of the year that further monetary policy stimulation was needed.

Six months later, in autumn 2014, after rate cuts totalling 75 basis points, the situation became critical as a result of the halving of the oil price in USD. This further dampened the already low inflation, although lower energy prices also benefit demand. On top of this, there were clear signals that the ECB was preparing a powerful stimulation package that would come into operation in spring 2015. Measures taken to improve the prospects for our trading partners are of course also positive for the Swedish economy, but if the negative interest rate differential between Sweden and the rest of the world were to increase, it could contribute to a strengthening of the krona. This would make it even more difficult for us to attain the inflation target.

At the beginning of 2015, the initial situation for the Riksbank was problematic. The risks of waiting to see what happened were assessed as greater than those connected to being proactive, as inflation had been low for a long time and inflation expectations were moving in the wrong direction. If monetary policy had not been made more expansionary, there was a risk that inflation expectations in the corporate and household sectors would fall further and I did not see this is a path to higher and more normal interest rates. The Executive Board was agreed that further stimulation was necessary. Starting from the February meeting, the repo rate was cut below zero and the first of a series of asset purchase programmes was begun. Furthermore, we signalled that there was a possibility of foreign exchange interventions if the krona appreciation already forecast were so strong and rapid that it jeopardised the upturn in inflation. We deliberately increased our focus on the development of short-run inflation, as we had a low tolerance of poor outcomes.

What considerations should govern our choice of future path?

We are now in April 2016 and I judge that we have largely succeeded in our efforts. Growth has become much higher than expected, although this is not merely due to monetary policy. The trend towards declining confidence in the inflation target has turned around in that inflation expectations have begun to rise again.

The monetary policy conducted is far from uncomplicated. In a world of minus rates it is more difficult to assess what impact the interest rate decisions will have on different channels, that is, how the transmission mechanism will work. It is also difficult to assess the stimulation effects of the bond purchases.² Moreover, the Riksbank has been clear that low interest rates increase indebtedness among households in a way that increases vulnerability in the Swedish economy. But most of all, I feel that our monetary policy is successful and better than the alternative. The inflation target would have risked losing its role as anchor for price-setting and wage-formation if we had not cut the repo rate in 2014 and 2015.

With regard to my thoughts on going forward, I would like to begin with the monetary policy meeting on 10 February. The report presented then stated that resource utilisation in Sweden was assessed to be at a normal level. The members of the Board agreed that the important thing for monetary policy is for the trend in inflation to come closer to the target and to avoid weakening confidence in the target. The upturn in inflation is expected to continue to be fitful and there may be surprises along the way. Whether, and if so how, monetary policy will react depends on the causes of such surprises and how they are deemed to influence the outlook for inflation. We were also agreed that we shall maintain a high level of preparedness for action, even between the ordinary meetings, in case further measures are needed to safeguard confidence in the inflation target.

Exactly how tolerant we are with regard to different outcomes may vary somewhat on an individual basis, which is natural on a committee. In my contribution to the debate at the meeting I wanted to say that I now have a greater tolerance for setbacks, in that we now have a more favourable situation in Sweden. I do not see this as advocating a new monetary policy strategy for the Riksbank. Instead I would describe it as a natural consequence of the change in the economic situation and inflation prospects. When the analyses of data point to resource utilisation normalising and inflation showing a trend towards the target, it is reasonable not to make monetary policy more expansionary, but instead to see developments in the slightly longer run. Or, as I have previously put it "it is time to take a longer view".

Some people probably think that central banks should always take a longer view, and normally I would agree with this. But I consider that the Riksbank was in a situation at the end of 2014/beginning of 2015 that required greater focus on the present, in words and deeds, to show that the Riksbank does not just talk, but also takes action to uphold confidence in the inflation target.³

As I said in other contexts, further monetary policy stimulation must entail advantages that are greater than the disadvantages. But I am not ruling out the possibility of further stimulation.⁴ I have already referred to the reason why this may be needed in my introduction today. The big questions, such as growth conditions, real interest rates and the behaviour of inflation in the short and long run, have in no way been answered. My point is that the global conditions for growth will gradually improve and that the course of inflation will normalise. But this depends on decisions made abroad, which govern the real interest rate that we in Sweden have to relate to.

Growth conditions in Sweden also depend on the Swedish economy's ability to constantly transform in an orderly manner. Monetary policy can function as a bridge over to a new normal situation. But the new normal will be determined by decisions that are beyond the central bank's control.

---