VoxEU Column Monetary Policy

Is the Riksbank neglecting the price-stability objective, counteracting full employment, and increasing household debt?

Sweden’s average inflation has consistently undershot its inflation target. This column argues that this has led to higher average unemployment and a higher household-debt ratio. The author, a former Deputy Governor of the Riksbank, argues that Sweden’s central bank is not fulfilling its mandate.

How do we know whether or not the Riksbank is neglecting the price-stability objective? Could it be that the Riksbank is not only neglecting the price-stability objective but is also counteracting the Riksdag’s and the Government’s full-employment objective as well as increasing household indebtedness?

The Riksbank’s mandate

What is the Riksbank’s mandate for monetary policy? The Sveriges Riksbank Act (1988) states: ”The objective of the Riksbank’s activities shall be to maintain price stability.” The Government Bill (Swedish Government 1997) about the Riksbank Act and how it shall be interpreted states: “As an authority under the Riksdag, the Riksbank shall also, without prejudice to the objective of price stability, support the objectives of the general economic policy with the aim to achieve sustainable growth and high employment.” (Italics added.)

  • The Riksbank has specified the price-stability objective as an inflation target of 2% for the annual rate of change of the Consumer Price Index (CPI).

High employment is harder to define. High employment must be interpreted as the highest sustainable employment, which is determined by the working of the economy and not by monetary policy. The highest sustainable employment is in practice (when the labour-market participation rate is independent of monetary policy) the same thing as the lowest sustainable unemployment.

  • Thus the mandate is to stabilise inflation around the inflation target and unemployment around the (lowest) long-run sustainable rate.
How do we know if the Riksbank has neglected the price-stability objective?

What is meant by “prejudice to the price stability”, that is, “neglecting” the price- stability objective, when the Riksbank has an inflation target of 2%? How do we know if the Riksbank is neglecting the price-stability objective or not?

Since the Riksbank does not have complete control over inflation, any deviation of inflation from the target cannot be due to neglecting the price-stability objective. The best the Riksbank can achieve is inflation fluctuating moderately around 2%.

  • This means that average inflation over a longer period becomes a criterion of whether the Riksbank is neglecting the price-stability objective. If average inflation over a longer period exceeds or undershoots the inflation target, the Riksbank is neglecting the price-stability objective.

However, one could argue that the Riksbank is not neglecting the price-stability objective as long as it is trying to bring inflation back to the target within a reasonable time.

  • The criterion could then be whether or not the Riksbank’s inflation forecast reaches the inflation target within a reasonable time.

But, as is shown by the National Institute of Economic Research (2013) (Figure 1), there are good reasons to doubt the Riksbank’s forecasts. Diagram 34 in the special study shows that the Riksbank’s forecasts for CPI inflation have systematically overestimated inflation and are therefore not reliable.

To check whether or not the forecasts reach the inflation target within a reasonable time is therefore not at good way to check whether or not the Riksbank is neglecting the price-stability objective.

Figure 1. Actual inflation vs. Riksbank forecasts

Sources: The Riksbanks, Statistics Sweden and NIER.

Comparing Sweden to other inflation targeters

In principle, it is of course doubtful to use forecasts or measures that the Riksbank has itself constructed to assess if the Riksbank is neglecting the price-stability objective. It is better to use measures and indices that some other institution, such as Statistics Sweden, has constructed.

Table 1. Average deviation from the inflation target

Sources: Reuters Ecowin and Statistics Sweden.

Average inflation over a longer period is easy to measure and, by comparing with other countries using inflation targeting, it is easy to see what is possible. Table 1 below (from table 1 in Svensson 2013b), shows the inflation target, what price index the target refers to, average inflation, and the average deviation from the target for a few countries that have had inflation targets as long as Sweden.

We see that Sweden stands out in comparison with the other countries. During 1997-2011, average CPI inflation has undershot the target by 0.6 percentage points. To this can be added that, during 2012-July 2013, average CPI inflation has only been 0.5%; thus undershooting the target by a full 1.5 percentage points.

  • With such a large deviation from the target compared to how other central banks have performed, it seems that one must conclude that the Riksbank is neglecting the price-stability objective, in particular when one observes how low average CPI inflation has been recently.
Canada and Sweden

Figure 2 below shows how a time series of the CPI for Sweden and Canada since 1997. The black line shows the hypothetical CPI increasing by 2% per year since 1997. The light blue line shows the CPI for Canada, and how closely if follows the black line. It is thus fully possible to keep average inflation close to 2% over a long period. The red line shows the Swedish CPI and how it on average has increased at a slower rate than 2%, especially in the last few years. If the CPI had increased by 2% on average since 1997, it would now have increased by almost 40% since 1997. It has actually increased by just a bit above 20%.

Figure 2. Consumer price indices: Canada v Sweden

Consequences of this neglect of the price-stability objective? Unemployment

As I have shown in Svensson (2013b) and is apparent in the Figure 3 below (Svensson 2013b, figure 11), 0.6 percentage points average inflation below the target has caused about 0.8 percentage points higher unemployment, about 38 000 jobs, during 1997-2011. That unemployment has been higher is because inflation expectations have equalled the target during this period. When actual inflation falls below expected inflation, unemployment becomes higher.

  • As shown in Svensson (2013d), currently unemployment may be as much as 1.2 percentage points higher (corresponding to about 60 000 jobs) compared to a situation where the Riksbank with more expansionary policy had kept inflation close to the target.

Figure 3. Unemployment and inflation

Consequences for household debt

As I have shown in Svensson (2013a) and discussed in a recent Vox column (Svensson 2013c), the Riksbank’s tight monetary policy has led to higher real household debt and a higher household debt ratio, measured relative to GDP or to disposable income. This is because the Riksbank’s policy has reduced the price level, nominal GDP and nominal disposable income faster than household nominal debt. Then real debt and the debt ratio increases.

One can show the long-run consequences for household debt in a different way. Figure 4 below shows that the CPI would have increased by about 23% since 2003, if average inflation had equalled the target. In fact, the CPI has only increased by 13%.

Figure 4. Sweden's consumer price index since 2003

Assume now that a house buyer in 2003 took out a mortgage of SEK 1 million. Assume for simplicity that the borrower has not amortised the debt but that ten years later it remains equal to SEK 1 million. Assume the borrower like most others was expecting 2% inflation per year for the next ten years. The black line in Figure 5 below (expressed in units of thousands of SEK) shows that the real value of the mortgage in 2003 prices would then have fallen to SEK 813,000 in 2013 (SEK 1million/1.23). But since the price level in reality has only increased by 14%, the real value of the mortgage has only fallen to SEK 878,000 (SEK 1 million/1.14).

  • This means that the real value of this debt is 8% higher than it would have been if the Riksbank had kept average inflation on target. The lower price level has also led to lower nominal GDP and lower nominal disposable income, so the debt ratio is also at least 8% higher; it is actually more than 8% higher, since the Riksbank policy has led to lower real GDP and real disposable income than if inflation had been on target.

Figure 5. Actual real value of a 2003 mortgage of SEK 1 million

Conclusions

The Riksbank has neglected the price-stability objective by allowing average inflation to undershoot the inflation target. This has led to higher average unemployment. The Riksbank has thereby not supported the objective of the general economic policy of high employment. Lower average inflation has also led to higher real household debt, and to a higher debt ratio relative to GDP and disposable income. All this in spite of the Riksbank justifying its tight policy by the presumption that the policy would lead to a lower debt ratio than an easier policy would.

All in all, the Riksbank is apparently not only neglecting the price-stability objective; it is actually both counteracting the high-employment objective of the Riksdag and the Government as well as increasing household indebtedness.

References

National Institute of Economic Research (2013), “The Riksbank Has Systematically Over-Estimated Inflation”, Special Study, The Swedish Economy August 2013, Stockholm

Svensson, Lars E O (2013a) “‘Leaning Against the Wind’ Leads to a Higher (Not Lower) Household Debt-to-GDP Ratio”, The Institute for Financial Research, Swedish House of Finance, Stockholm School of Economics.

Svensson, Lars E O (2013b), “The Possible Unemployment Cost of Average Inflation below a Credible Target”, The Institute for Financial Research, Swedish House of Finance, Stockholm School of Economics.

Svensson, Lars E O (2013c), “The Riksbank is Wrong about the Debt: Higher Policy Rates Increase Rather Than Decrease the Household-Debt Ratio”, VoxEU.org, 4 September.

Svensson, Lars E O (2013d), “Some Lessons from Six Years of Practical Inflation Targeting”, The Institute for Financial Research, Swedish House of Finance, Stockholm School of Economics.

Sveriges Riksbank Act (1988), Sveriges Riksbank Act (1988:135) in its wording as of 1 July 2011, Sveriges Riksbank.

Swedish Government (1997), Riksbankens ställning (The position of the Riksbank) (in Swedish), Government Bill 1997/98:40, Stockholm.

2,730 Reads