Monetary policy of the last two years in the rear-view mirror, or what we managed to prevent
In late June 2021, the Czech National Bank started to increase its main policy rate (the two-week repo rate), which had been at the very low level of 0.25% for more than a year since the start of the coronavirus pandemic in the Czech Republic in May 2020. The rate went up quickly in the months that followed, reaching its present level of 7% in a series of nine steps.
This blog article retrospectively evaluates the appropriateness of monetary policy and in particular quantifies the effect of the interest rate increases made by the Czech National Bank (CNB) between the third quarter of 2021 and the present. The results of the analyses described below suggest that the past monetary tightening was entirely appropriate in light of the approaching strongly inflationary economic developments, although with the benefit of hindsight it should have happened sooner and faster. We then draw up a hypothetical scenario of no interest rate increases to quantify the effects that extremely easy monetary policy might have had in the past year. We also formulate a hypothetical simulation reflecting the potential loss of credibility of the inflation targeting regime and unanchoring of medium-term inflation expectations from the 2% target in an environment of runaway inflation under the assumption of passive monetary policy.
Tools usually used to assess MP appropriateness
At least once a year, the CNB’s Monetary Department retrospectively assesses the fulfilment of its past forecasts, including those for the main macroeconomic indicators such as inflation, the koruna exchange rate and interest rates. The results of the most recent assessment exercise, covering the 2020 forecasts, were published as an appendix to the spring 2022 Monetary Policy Report. This approach has the advantage of being comprehensive and of identifying sources of deviations between the historical forecast under assessment and the observed outcomes from the perspective of the g3+ core prediction model. However, the time lag needed to conduct this exercise is something of a disadvantage: to be able to determine how inflation evolved in reality over the monetary policy horizon at the time, we need at least 18 months to have passed since the creation of the forecast under assessment. The results of this exercise for the 2021 forecasts, on the basis of which the current rate-increase cycle was started, will thus not be available until late 2022/early 2023. The forecasts from the first half of this year will be assessed in a study prepared yet another year later.
As an alternative, we evaluate the appropriateness of CNB monetary policy based on the results of a different exercise performed quarterly by the Monetary Department for each Monetary Policy Report: the assessment of the fulfilment of the inflation target. This takes the form of an analysis identifying observed sources of deviations of monetary policy-relevant inflation from the 2% target, which is a regular part of the Chartbook in the Monetary Policy Report (Chart E.15) and is also described in a Box in the same publication once a year. One of the target-fulfilment factors considered in this approach is the effect of monetary policy, that is, the explicit contribution of historically observed interest rates in addition to the effect of the exchange rate. If the effect of interest rates counterbalances the effects of other shocks to inflation in the period concerned (taking into account the monetary policy transmission lag), the previous monetary policy can be considered to have been appropriate. By contrast, if monetary policy contributes to inflation deviating upwards (downwards) from the target, it can be said that with the benefit of hindsight the previous monetary policy should have been tighter (looser) and interest rates should have been higher (lower). The advantage of this approach is that we can use the core prediction model to break down the deviation of inflation from the 2% target into the contributions of individual factors as soon as we have data on all the observed variables available for the quarter under assessment. Figure 1 shows the most recent available breakdown of the deviation of monetary policy-relevant from the 2% target taken from the spring 2002 Monetary Policy Report. According to this chart, with hindsight, and taking into account the monetary policy transmission lag, monetary policy was too easy in 2020, because in 2021 it started contributing increasingly to accelerating inflation, which, due to the effects of monetary policy and all other relevant factors, was well above the target and diverging rapidly from it.
Figure 1 – Deviation of monetary policy-relevant from the 2% target
Inflation (year-on-year changes in %, contributions in percentage points)
Another possible approach to assessing the appropriateness of previous monetary policy is to construct hypothetical counterfactual scenarios that differ from the observed outcomes in certain predefined respects.
Counterfactual scenario
To draw up this scenario we will use the g3+ core prediction model and try to answer the question of how the economy would probably have evolved had interest rates not been increased at all in the last year and had rates stayed at the low level seen at the start of the pandemic. Such a scenario thus simultaneously gives a model-consistent (and model-conditional) answer to the question of how strong a braking effect the actual monetary policy tightening had on inflation.
We therefore prepare a hypothetical scenario in which – ceteris paribus – the CNB’s main policy rate would have stayed at the initial level of 0.25% from 2021 Q3 through to 2022 Q2.[1] We will assume that the central bank would have openly communicated this policy in advance.[2] We will also assume in this simulation that by means of its communications the central bank would simultaneously have succeeded in maintaining full credibility in the inflation targeting regime and keeping medium-term inflation expectations anchored at 2%. In the model simulation, we will thus move interest rates back to their initial 2021 Q2 level.
In such case, the Czech economy, which was already under inflationary pressure, would have additionally been exposed to strongly expansionary monetary policy lasting for the entire period up to the present. The hypothetical lower interest rate differential against the Eurozone would have been reflected in the exchange rate of the koruna, which would have been CZK 4 weaker against the euro than we observed in reality. The substantially easier monetary conditions would have resulted in a gradually widening gap between simulated and actually observed inflation, which would have reached around 6 percentage points (pp) by 2022 Q2. Owing to the lag in monetary policy transmission, the impacts would have strengthened further over time, peaking in 2022 Q3 with inflation more than 7 pp higher than it was in reality. The loose monetary policy would also have fostered growth in household consumption and overall GDP last year and especially this year. The simulation results are shown in Figure 2 (red line).
Figure 2 – Counterfactual simulation
(Un)anchoring of inflation expectations
Over and above the previous version of the counterfactual simulation, it is natural to ask what would have happened to the anchoring of inflation expectations to the 2% target given such hypothetically extremely and inappropriately easy monetary policy. If the central bank, faced with extreme inflation pressures,[3] had remained entirely passive and had de facto given up on achieving the inflation target in the period ahead, it is highly likely that medium-term inflation expectations would indeed have become unanchored. Along with the baseline counterfactual scenario, we therefore present the results of a simulation that additionally considers a 2 pp increase in inflation expectations from 2021 Q3 in Figure 2 (yellow line).
Had the central bank remained passive even in this case, its monetary policy would de facto have been even easier and the resulting inflationary effects even stronger. Inflation would have peaked 13 percentage points higher than it did in reality.[4] The exchange rate of the koruna against the euro would meanwhile have been 5 CZK/EUR weaker. The impacts on growth in household consumption and GDP last year and this year would also have been rather stronger than in the baseline counterfactual simulation. However, these effects are small relative to the additional increase in the inflation rate and would hardly have counterbalanced the perceived decline in the people’s purchasing power and the substantial drop in the value of their savings.
It can also be assumed that in the case of a longer-lasting monetary policy non-reaction (as assumed in this scenario) the growth in inflation expectations in the high-inflation environment might ultimately have been even greater. This would suggest even stronger impacts on inflation (and an even greater subsequent tightening to correct inflation expectations and steer them back to the 2% target).
Conclusion
The results of our counterfactual simulation are consistent with the conclusions of previous regular analyses. With the benefit of hindsight, and with knowledge of the strength of the inflationary pressures, monetary policy was too easy in 2020 and the first half of 2021. If, during the pandemic, the CNB had known the exact enormity of the inflationary shocks that the global and Czech economies were subsequently going to face, the appropriate response would have been to start tightening rapidly in autumn 2020.[5] This would have ensured that inflation did not diverge from the target as significantly as we have seen in recent quarters.
If, hypothetically, the CNB had moreover decided for some reason to openly abandon the active use of interest rates as a monetary policy instrument in the past year, the situation today would have been even more dramatic – inflation would have been running at around 23% and the exchange rate of the koruna would have been around CZK 28.50 to the euro. And if inflation expectations had additionally become simultaneously unanchored from the target, we would have observed an inflation rate of more than 28% and the koruna at nearly CZK 30 to the euro.
[1] We also assume a constant repo premium of 0.15 pp. This implies that the 3M PRIBOR market rate stays at 0.4%.
[2] For instance following the example of the ECB, which until recently assessed the emerging wave of inflation as temporary and connected with previous Covid restrictions, post-Covid opening of economies and negative supply/cost shocks and hence as not requiring any monetary policy response.
[3] The central bank was initially faced primarily with increased inflation pressures in the domestic economy. However, these were later increasingly amplified by foreign price shocks of an unprecedented scale.
[4] The impacts of growth in prices on households are always individual to some extent, depending on the specific composition of their consumer basket. When the average inflation rate in the economy as a whole is low and close to the inflation target, the differences are negligible. In an environment of runaway inflation, however, the impacts on individual households’ budgets are highly differentiated. Some categories of households could thus be exposed to an even greater increase in the cost of their consumer basket relative to the average for the economy as a whole. The impacts on the finances of individual firms could also differ similarly widely in an environment of runaway inflation, as there could be sizeable changes in consumer behaviour and spillovers of demand between firms and industries.
[5] The question is whether the CNB would have found the courage to implement such a tightening at the height of the coronavirus pandemic in winter 2020/2021, when 200–250 people a day were dying with COVID-19 despite the heroic efforts of front-line workers, and a state of emergency and a complete lockdown (including a curfew and a ban on inter-municipality travel) had been declared in the country.
The opinions expressed in this article are those of the authors and do not necessarily reflect the official position of the Czech National Bank.