Stefan Elfwing sent me this report from the Swedish Riksbank:
Deputy Governor Lars E.O. Svensson stated that he enters a reservation against parts of the draft Monetary Policy Report. Mr Svensson prefers that the repo rate is cut by 25 basis points to 1.75 per cent at this meeting and then a lower repo-rate path which is at 1.5 per cent from the first quarter of 2012 and until the end of the first quarter of 2013, and which then rises at an even pace to slightly above 3 per cent at the end of the forecast period.
He feels that the repo-rate path in the main scenario is too high for three reasons. Firstly, as he has said, an unjustified high forecast for policy rates abroad gives a bias towards a too high repo-rate path. Secondly, there is a bias towards overestimating resource utilisation. Thirdly, even if one accepts the high forecast for policy rates abroad and the high estimate of resource utilisation and the sustainable unemployment rate of 6.5 per cent, one can still show that a lower repo-rate path stabilises CPIF inflation better around 2 per cent and unemployment better around a level of 6.5 per cent. Regardless of how one views it, the main scenario thus entails monetary policy that is not well balanced.
Mr Svensson said that he actually thinks that this is fairly self-evident. Since the meeting in September the situation for the Swedish economy has deteriorated. The forecasts for inflation and resource utilisation have been adjusted downwards for an unchanged repo rate path. The repo-rate path in the main scenario has been lowered in relation to September, but not far enough. With the new repo-rate path the forecast for CPIF inflation is still too low and the forecast for unemployment is still too high. Lowering the repo-rate path further would improve target fulfilment for both inflation and unemployment. This should be obvious to everyone. Mr Svensson felt that one doesn’t need to be an expert on monetary policy to understand it. (Emphasis added throughout this post)
Keep in mind that Sweden is an ultra-nice society. You might want to translate the last sentence into Italian, and imagine how it would have been phrased if uttered by one of Berlusconi’s henchmen. I found the entire document a joy to read, as Svensson continually used his razor sharp reasoning to slice and dice his colleague’s sloppy arguments for doing nothing.
So why isn’t the repo-rate path being lowered further? Mr Svensson went on to say that a few muddled reasons are given at the end of Chapter 2. It is claimed that economic activity will fall because companies and households will postpone consumption and investment due to the uncertainty over future developments. In such a situation, it is doubtful whether monetary policy will have the same immediate effect as indicated by the historically-estimated links. It would require a more uneven repo-rate path to stimulate the economy.
This appears to be speculation and has no foundation. The arguments are not convincing and raise several questions. Why is it increased uncertainty and not lower expectations of future income and demand that will make households and companies reduce their consumption and investment? And if it is increased uncertainty, why should this mean that monetary policy has less impact? And why should a specially-uneven repo-rate path be necessary in this situation? Why wouldn’t a lower, relatively smooth repo-rate path provide a better and acceptable outcome? And how can one discuss these issues without showing how the various repo-rate paths would look? Where is the analysis that supports these claims? As far as Mr Svensson could see there was no such analysis. It all gives an impression of excuses added on afterwards.
This also drives me nuts. Tight money reduces AD growth expectations. That naturally makes people more uncertain about the future. Then my opponents claim the problem isn’t tight money, it’s uncertainty. Yes, and panic on the Titanic wasn’t caused by a iceberg gouging the ship, it was caused by uncertainty about when they’d reach New York.
If, in this situation, the repo rate were really to have less effect on inflation and resource utilisation than usual, it would be a reason to cut the repo rate further, not less, to have the same effect.
Memo to Richard “pushing on a string” Fisher. Please read Lars Svensson.
In autumn 2008 and later, following Lehman, in a situation with maximum uncertainty and minimum confidence among economic agents, it was by acting forcefully and making large interest-rate cuts that enabled central banks around the world to contribute to ensuring that the crisis did not develop into the Great Depression II.
If one takes the increased uncertainty seriously, it must be so that the increased uncertainty means a greater probability that the situation may deteriorate so much that the zero lower bound may bind the policy rate in the future. In this case, academic research results unequivocally say that there is justification for lowering the policy rate further and more quickly than is justified by the forecasts for inflation and resource utilisation. This also reduces the risk of being caught in the future in a situation with a binding lower bound for the policy rate.
Obviously I think he’s being too kind to the other central banks. Note that as early as the spring of 2008 there was a real danger of a liquidity trap developing in America. Yet the Fed never cut rates once between May and October, even after Lehman failed. But the basic argument is right, if there’s downside danger of a zero rate trap, you move aggressively, you don’t wait for more information.
Moreover, it is stated in the draft Monetary Policy Report that a lower repo-rate path would entail a risk that housing prices and household indebtedness once again began to develop in a way that would lead to imbalances in the economy in the longer run. Mr Svensson wondered what has happened to the conclusions from the Riksbank’s large inquiry into risks in the housing market. This stated that the policy rate is a blunt instrument, not suitable for influencing housing prices and indebtedness as it can have very negative consequences for the real economy, and that there are other instruments, such as the mortgage ceiling and so on, that are much more effective. And what kind of imprecise and indeterminate imbalances are we talking about? The results of the Riksbank’s inquiry show that there are no signs that houses are over-valued, but that housing prices are quite compatible with the fundamentals, mainly a large demand and small supply.
. . .
Moreover, this talk of unspecified imbalances in the longer run appears to mean that all of the work on financial stability is repudiated. The Financial Stability Report closely monitors threats to financial stability and risks due to high leverage and low liquidity are managed much better with capital adequacy and liquidity requirements than with the repo rate.
So, why should housing prices and household borrowing be promoted to target variables that justify inflation being too low and unemployment being too high? Monetary policy should not be the first, but the last line of defence with regard to financial stability, to be used only in cases where the normal financial stability tools do not work or are insufficient. There is now an abundance of instruments for managing financial stability.
Nor should the repo rate be an explicit or implicit target variable. It should solely be used as a means to stabilise the correct target variables, inflation and resource utilisation.
As Governor Strong said before he died in 1928:
Must we accept parenthood for every economic development in the country? That is a hard thing for us to do. We would have a large family of children. Every time one of them misbehaved, we might have to spank them all.
Exactly. Deal with financial excess through better financial regulation, and deal with inflation and output (NGDP) via monetary policy. Back to Svensson:
It is clear that the repo-rate path in the main scenario gives a CPIF forecast well under 2 per cent. A lower interest-rate path abroad, together with the repo-rate path in the main scenario, gives a greater interest rate differential and a stronger krona that brings down inflation. The lower repo-rate path gives higher CPIF inflation closer to the target and a much lower forecast for unemployment closer to 5.5 per cent. Target fulfilment for both inflation and unemployment will be much better with the lower repo-rate path.
Note that this result is not sensitive to the assumption of a sustainable unemployment rate of 5.5 per cent. The lower interest-rate path also gives better target fulfilment if the sustainable unemployment rate is assumed to be 6 or 6.5 per cent.
Imagine that, a central bank refrains from additional stimulus even though it would push both inflation and employment closer to target. Lucky our Fed never misses those free lunches, those $100 bills lying on the sidewalk.
Here Svensson lectures his colleague on the difference between a policy indicator and a policy target:
Deputy Governor Lars E.O. Svensson said that he reacted when Mr Ã–berg said it was unclear what the high unemployment entails for resource utilisation, immediately after Mr Svensson had said that the unemployment gap was the most relevant and reliable measure of resource utilisation as a target variable. Here it is important to distinguish between measures of resource utilisation as a target variable and measures of resource utilisation as indicators of inflationary pressures. They are not the same thing. This is something that has been discussed several times at earlier meetings, said Mr Svensson.
Mr Svensson said that Mr Ã–berg was probably referring to the measures of resource utilisation as indicators of inflationary pressures. Such measures enter monetary policy and influence the repo-rate path solely through their effect on the inflation forecast. They are merely indicators, not target variables. Measures of resource utilisation as a target variable are a different thing. The unemployment gap between actual unemployment and the sustainable unemployment rate is a target variable. All else being equal, it is desirable to stabilise unemployment around the sustainable level. This unemployment gap is not necessarily the best indicator of inflationary pressures.
And here he discusses “loss of face:”
With reference to Ms Ekholm’s contribution to the discussion on the risk of the Riksbank causing damage to its reputation if it first cuts the repo rate and then needs to increase it again, Mr Svensson felt that it was quite possibly this type of consideration regarding loss of face that lay behind the discussion of an uneven repo-rate path in the Monetary Policy Report. However, it entails a greater risk of damage to its reputation for the Riksbank to deliberately choose an interest-rate path that does not entail a well-balanced monetary policy with regard to inflation and resource utilisation. At each monetary policy meeting one should draw up a repo-rate path that best stabilises both inflation and resource utilisation in an unbiased sense. This ought to be what is best for the Riksbank’s reputation in the long run. The probability should then normally, specifically using symmetrical probability distributions, be roughly the same for later needing to raise or lower the repo-rate path. These changes in the path need not in themselves lead to any damage to the Riksbank’s reputation, according to Mr Svensson.
Loss of face? This is the Riksbank, not the PBoC! I’m only 1/4th Scandinavian, and I don’t worry about losing face. Recall the Fed meeting in late 1937 that Marcus Nunes dug up? A Fed official basically admitted that they should ease, but failed to do so because he was fearful it would make their previous decision (to raise reserve requirements) look bad. I also like the symmetry argument. Does anyone believe the Fed is equally likely to ease further or tighten further over the next 6 months?
And here Svensson basically says; “Thank God the markets don’t believe we’ll be as tight at my colleagues predict:”
The red curve in Figure 2 shows how the Swedish yield curve would look if the repo-rate path in the main scenario gained full credibility and forward premiums are normal. It is then clear that a five-year rate would be roughly 190 basis points higher than now. This would not be good for the Swedish economy. It gives reason to really hope that the market continues to set the current low market rates, emphasised Mr Svensson.
And here he explains what monetary policy should and should not do:
Deputy Governor Lars E.O. Svensson commented on Mr Nyberg’s discussion of macroprudential supervision, that is, financial stability policy, by emphasising that it is important to realise that policy for financial stability is not the same as monetary policy.
Fiscal policy is not considered to be monetary policy. Fiscal policy has its objectives, primarily efficiency, stability and an even income distribution, and its instruments, primarily taxation and spending. Monetary policy has its objectives, stable inflation and resource utilisation, and its instruments, primarily the policy rate and communication. Fiscal policy influences inflation and resource utilisation. This means that monetary policy must give consideration to the way that fiscal policy is conducted when setting the interest rate, but it does not mean that fiscal policy is monetary policy.
I hope all MMTers will read this report.
In the same way, the policy for financial stability and monetary policy are different policies. The policy for financial stability has its own objective, that is financial stability, and its own instruments, primarily regulation and supervision. The policy for financial stability affects financial markets, spreads between interest rates, the functioning of financial markets and the transmission mechanism. This means that monetary policy must give consideration to how the policy for financial stability is conducted when setting the interest rate to attain the monetary policy objectives. It also means that the policy for financial stability must take into account how monetary policy is conducted when supervision and regulation are used to attain and maintain financial stability. But it does not mean that monetary policy and the policy for financial stability are the same thing.
I hope all Austrians will read this report.
And finally, Lars Svensson explains that it’s dangerous to base policies on policymaker predictions of future market outcomes that are wildly at variance with market predictions of future market outcomes (skim the first three paragraphs):
Deputy Governor Lars E.O. Svensson pointed out that he has a more pessimistic view of the real economy than the one presented in the draft Monetary Policy Report with regard to inflation and GDP abroad. He fears that it is less likely that the debt crisis in Europe will be resolved in a good way and that the effects on the real economy and fiscal policy tightening will thus be more negative. As he has pointed out earlier, and as Ms Ekholm has stressed, there are problems with the assumptions regarding interest rates abroad and the forecast for policy rates abroad. As he had demonstrated with the aid of a figure at the monetary policy meeting in July, the forecast for policy rates abroad has systematically been too high for several years, with the outcome systematically lower than the forecast. A too high forecast for policy rates abroad leads to a bias towards a too high repo-rate path, all else being equal. A higher interest rate path abroad will, all else being equal, lead to a smaller difference between Swedish interest rates and those abroad, and to a weaker krona. The Swedish repo-rate path must then be higher to counteract this. [Note; he means it will erroneously be set too high.]
Mr Svensson then showed some figures to illustrate his reasoning. The figures contain forecasts and assessments beyond the normal forecast horizon that Mr Svensson has made himself. In Figure 1 the yellow line shows the current forecast for TCW-weighted interest rates abroad. The grey line shows TCW-weighted implied forward rates, adjusted by normal forward premiums, that is, 1 basis point per month. The forecast is very high above the implied forward rates and gives rise to a substantial upward shift in the Riksbank’s repo-rate path. There is no discussion of these important circumstances in the draft Monetary Policy Report and there is no explanation of this high forecast, despite the fact that it has major consequences for the repo-rate path.
The forecast for interest rates abroad can also be perceived as an assumption on and a forecast for yield curves abroad, that is, interest rates abroad for different maturities. In Figure 2 the grey curve shows the actual TCW-weighted yield curve abroad, more specifically that which is compatible with the extended forward rate curve in Figure 1, while the yellow curve shows the yield curve abroad that is compatible with the Riksbank’s forecast for policy rates abroad. A TCW-weighted five-year market rate is roughly 110 basis points. The Riksbank’s high forecast for policy rates abroad corresponds to a five-year rate that is 100 basis points higher. Mr Svensson pointed out that one could simplify by saying that the Riksbank’s analysis is based on the five-year rates abroad being 100 basis points higher than they actually are.
Mr Svensson pointed out that in this context it may be interesting to see what Norges Bank had forecast for forward rates abroad in its most recent Monetary Policy Report at its meeting on 19 October. They stated that there were no strong reasons for assuming that interest rates abroad would normalise more quickly than is indicated by market rates, that is, by forward rates. They also emphasised that the Federal Reserve had signalled that its policy rate will remain low for a long time. Norges Bank’s forecast for policy rates abroad thus agrees with forward rates.
To sum up, Mr Svensson claimed that the forecast for policy rates abroad is much too high and that this contributes to the repo-rate path being too high. It would be better to let the forecast to be based on forward rates abroad and then adjust it on the basis of an assessment of monetary policy in the euro area, the United Kingdom and the United States. Mr Svensson claimed that he had pointed out on several earlier occasions that it is important to monitor this issue, and that potential deviations from implied forward rates should be justified and carefully discussed, as they have a major effect on the domestic interest-rate path.
Translation into Swedish. “My God! How many times do I have to explain this? Even those country bumpkins in Norway get it.”
Lars Svensson, market Keynesian. If only we could convince him to become a market monetarist.