Comparing Sweden & Norway on inflation
Paul Krugman has a post talking about how tighter monetary policy caused Sweden to go into deflation. Here is the chart showing their deflation.
OK… we can see the deflation there far to the right. Paul Krugman talked about Sweden raising the interest rate to battle bubbles. Let’s look at the interest rate in Sweden.
Yes, they certainly raised the interest rate. But didn’t Norway also raise their interest rate? Norway also has its own currency.
Norway actually raised their interest rate higher and longer than Sweden. So I guess we would see deflation in Norway, right?
Hmmmm, Norway has an inflation rate of 2.6%. OK, Mr. Krugman, what am I missing here? Your logic is not solid. Where is the answer?… I compare consumer spending in these two countries.
Swedish consumer spending rose 13% since 2009. What about Norway?
In Norway, consumer spending rose 17% since 2009. That would help inflation if people were spending more. Maybe inflation is stronger in Norway over Sweden due to more purchasing power of people, instead of what the central bank interest rate is.
“Paul Krugman has a post talking about how tighter monetary policy caused Sweden to go into deflation. Here is the chart showing their deflation.
[Graph]
OK… we can see the deflation there far to the right. Paul Krugman talked about Sweden raising the interest rate to battle bubbles. Let’s look at the interest rate in Sweden.
[Graph]
Yes, they certainly raised the interest rate. But didn’t Norway also raise their interest rate? Norway also has its own currency.
[Graph]
Norway actually raised their interest rate higher and longer than Sweden. So I guess we would see deflation in Norway, right?
[Graph]
Hmmmm, Norway has an inflation rate of 2.6%. OK, Mr. Krugman, what am I missing here? Your logic is not solid…”
Paul Krugman:
“But of course its own former deputy governor — and my former colleague — Lars Svensson, more or less frantically warned that the Riksbank was making a terrible mistake by tightening money despite low inflation and lots of economic slack.”
How much economic slack is there in Sweden in Norway? The easiest way to measure this is by looking at the unemployment rate:
https://research.stlouisfed.org/fred2/graph/?graph_id=173342
Norway raised the key policy rate from 1.25% in October 2009 to 2.00% by May 2010 and later to 2.25% in May 2011, or an increase of 1.25 points over a period of 19 months. During that time the unemployment rate ranged from 3.2% to 3.7%.
In contrast Sweden raised the repo rate from 0.25% in July 2009 to 2.00% by July 2010, or 1.75 points over a period of 12 months. During that time the unemployment rate ranged from 7.7% to 8.7%.
Even when one takes into account the different measures of NAIRU the slack was still much greater in Sweden than in Norway. According to OECD estimates NAIRU was 7.2% in Sweden in 2009 and 2010 falling to 7.0% from 2011 on forward. NAIRU was 3.3% in Norway throughout. Thus unemployment was as much as 1.5 points higher than NAIRU when the repo rate was increased in Sweden but it was actually 0.1 points less than NAIRU during the first key policy rate increase in Norway.
I think one also needs to keep in mind that the official inflation rate targets of Sweden and Norway are 2.0% and 2.5% respectively. During the time they raised their policy rates year on year core inflation ranged from 0.9% to 2.4% in Norway and from 0.8% to 1.2% in Sweden:
https://research.stlouisfed.org/fred2/graph/?graph_id=173345
So both countries were missing their target inflation rates, but Sweden was usually missing its target rate by a larger margin.
So when you combine that with the larger degree of slack, and the fact that the policy rate was increased by a larger margin in a shorter amount of time, Sweden’s monetary policy certainly looks much tighter that Norway’s during the period of interest rate increases.
To see why interest rate policy has resulted in accelerating inflation in Norway and deflation in Sweden it might be useful to look at some Taylor Rules.
The following are the 1999 version of the Taylor Rule with twice the weight on output gap as the inflation target gap. They take into account that Sweden’s inflation target (2.0%) is different from Norway’s (2.5%) and use core inflation. The output gap is measured using a post-1995 Okun’s Law coefficient from Laurence Ball’s paper “Okun’s Law: Fit at Fifty?” and the gap between the OECD NAIRU and the harmonized unemployment rate.
Here is Sweden first. The blue line is the 3-month interbank rate and the red line is the Taylor Rule. (Sweden’s 3-month interbank rate is slightly below the repo rate.):
https://research.stlouisfed.org/fred2/graph/?graph_id=173348&category_id=
Note that in July through December 2010 when Sweden did its first four interest rate increases, the Taylor Rule prescribed a negative policy rate. And in January through July 2011 during the remaining three interest rate increases the Taylor Rule continued to prescribe a rate over a point below the rate adopted by the central bank.
Since May 2012 the Taylor Rule has always been negative and the gap between it and the actual policy rate grew much larger through the middle of 2013. Consequently unemployment rose from 7.5% in April to 8.4% in November 2012, and Sweden’s unemployment rate has been consistently above its NAIRU of 7.0%-7.2% since February 2009.
In short, interest rate policy has been consistently too tight since Sweden ended QE and left the zero lower bound in July 2010. Thus it’s no surprise what had been one of the best performing economies in the EU in 2009-2010 is now entering full blown deflation.
Now let’s take a look at Norway. The blue line is the 3-month interbank rate and the red line is the Taylor Rule. (Norway’s 3-month interbank rate is slightly above the key policy rate.):
https://research.stlouisfed.org/fred2/graph/?graph_id=173347
Note that in October and December 2009, when Norway did its first two interest rate increases the Taylor Rule prescribed a policy rate even higher than what was adopted. Only in the case of the May 2010 and May 2011 interest rate increase was the Taylor Rule below the actual policy rate, and never by more than 0.9 points at the time of the increase.
And since January 2012, apart from November 2012 through May 2013, a period of 7 months, the Taylor Rule has almost always been above the actual policy rate. This is partly attributable to the fact that Norway’s unemployment rate was at or below its 3.3% NAIRU for the entire 12 month period from November 2011 through October 2012, dropping as low as 3.0% in March and April 2012.
Thus the period of tightening was much less of a policy error in Norway’s case than Sweden’s. And for over two years now Norway’s monetary policy has been generally on the expansionary side. It’s no wonder that Norway’s inflation rate has accelerated to the point that it is now roughly on target.
Mark,
The unemployment rate in Sweden normally spikes up after Christmas, so you can’t use that 8.7% rate as showing slack. It was expected and normal.
Edward,
Thanks to your comment I noticed a mistake.
“In contrast Sweden raised the repo rate from 0.25% in July 2009 to 2.00% by July 2010, or 1.75 points over a period of 12 months. During that time the unemployment rate ranged from 7.7% to 8.7%.”
should read
“In contrast Sweden raised the repo rate from 0.25% in July 2010 to 2.00% by July 2011, or 1.75 points over a period of 12 months. During that time the unemployment rate ranged from 7.7% to 8.7%.”
First of all the month that unemployment was 8.7% was July 2010, which was the month that interest rates were first raised. (No, Christmas does not occur in July.)
Furthermore these unemployment rates are all seasonally adjusted so it is absolutely false that the unemployment rate in Sweden normally spikes up after Christmas.
Mark S,
You don’t understand the logic here. Whatever the slack if interest rates were important for inflation a 2.5% move would have the same sign effect in both countries, with possibly differing magnitudes. But in one inflation went down in the other it went up. Conclusion: there are forces at work here which are much bigger than a 2.5% rate hike.