The fight against inflation has largely been won in the euro area, after a sharp but short-lived inflation episode over 2021-2023. This column argues that energy prices can explain the dynamics of inflation over the recent period better than the interest rate policy of the ECB. At the same time, higher interest rates led to significant losses incurred by central banks as they had to make interest transfers to banks holding reserves. A tiered system for remuneration of bank reserves would have allowed for a significant reduction in interest transfers to the banks, while keeping the central bank’s operating procedur unchanged.
It appears that the fight against inflation has been won in the major industrialised countries. We show this in Figure 1 for the euro area and the US. After a sudden surge during 2021-22, inflation dropped precipitously during 2022-23. Since then, inflation has hovered between 2 and 3%. Compared to the inflation experience during the 1970s and 1980s, this is a remarkably short-lived inflation episode. Thus, barring new shocks (the US tariff shock comes to mind), one can conclude that the euro area and the US have been quite successful in eliminating inflation.
In this column we ask two questions. First, what is the contribution of central banks in this success? Second, what have been the costs of this success? We concentrate on the euro area.
Figure 1 Inflation in the euro area and the US
Source: Bureau of Labor Statistics and European Central Bank
How successful was the ECB in fighting inflation?
We start from Figure 2. This shows the rate of inflation in the euro area since 2021 and the policy rate applied by the ECB. The latter is in fact the deposit rate, i.e. the rate of remuneration on bank reserves.
Figure 2 Inflation and interest rates in the euro area
Source: European Central Bank
We observe the following from Figure 2.
The ECB started raising the deposit rate when inflation was at its peak in September 2022. It took a year for the deposit rate to reach a peak, in October 2023. At that point inflation has dropped to 2.9% (from a peak of 10.6% in October 2022).
Can this swift decline in the rate of inflation during this one-year period be caused by the steep increase in the deposit rate engineered by the ECB during the same period? The answer is that this is highly implausible. We know from the empirical literature that there are long lags between interest rate changes and inflation. This empirical evidence suggests that interest rate changes have significant effects on inflation only after one year or more (e.g. Favero 2001, Jorda, et al. 2020, Willems 2020, Aruoba and Drechsel 2024, Burr and Willems 2024). Aruoba and Drechsel (2024) even find lags of three years. The existence of long lags in the transmission of monetary policy on the price level is recognised by the ECB in a blog on its website when it writes that “The transmission mechanism is characterised by long, variable and uncertain time lags” (ECB 2025) thereby echoing Friedman’s well-known conclusion that monetary policies affect the economy with “long and variable lags” (Friedman 1961).
Thus, the interest rate hikes during 2022-23 could have affected inflation significantly only from the end of 2023 on. But at that moment observed inflation had already dropped below 3%. This must have been caused by variables other than the interest rate. We return to this.
From September 2023 until June 2024 the ECB kept the deposit rate at 4% (for almost a year) and nothing happened to inflation. That is when one would have expected the interest rate hike to have started to lower inflation. Nothing happened to inflation, which remained mostly above 2%.
In June 2024 the ECB started lowering the interest rate, but inflation remained stuck above 2%. Something else must have been going on, unrelated to the ECB’s interest rate policy.
The other variable that is more likely to have affected the rate of inflation is the index of fuel prices. We show this in Figure 3. This presents the rate of inflation in the euro area together with the monthly observations of the year-on-year increases in the fuel price index (as computed by the International Monetary Fund). We find that the fuel price increases, although quite volatile, precede the movements in inflation by six to 12 months throughout the period: Around March 2021, energy prices, especially gas prices, started a process of spectacular increase. Approximately six months later, this induced a sharp increase in inflation. The energy price increases remained elevated for approximately two years. In March 2023 they started a steep descent. After six months, inflation started a steep decline until it reached 2.4% in November 2023. From then on inflation has not declined further. This stalling of the rate of inflation coincides with a renewed (but limited) increase in the energy price index.
Figure 3 Inflation and energy price increases
Source: inflation, ECB; Fuel price index (yearly change), IMF
Thus, it appears that there is a strong association between energy price changes and subsequent inflation. The episode of the sharp decline in the rate of inflation (from October 2022 until October 2023) is particularly revealing. This decline was preceded (by approximately six months) by a steep decline in energy prices. It is this decline in energy prices that is most likely to explain the steep decline in inflation, and not the increase in the interest rate that was engineered by the ECB during the same period.
Another feature of Figure 3 is noteworthy. We observe that from October 2023, the decline in inflation stalled even though this is the moment when the interest rate hikes might have started to work. The reason for this stalling can be found in the movements of the energy price index. The latter started to increase slightly at the end of 2023, thwarting the efforts of the ECB to reduce inflation further by maintaining high interest rates.
We conclude that the cycle in inflation is well explained by the cycle in energy prices (for a similar conclusion see Arce et al. 2023). The interest rate policies of the ECB do not seem to have had much influence on this inflation cycle. One may even wonder whether in the absence of interest rate hikes by the ECB the rate of inflation would have been pretty much the same. If this is so, the European Central Bank was irrelevant, and it may as well not have increased the interest rate.
One can object to this conclusion that the spectacle of a central bank not reacting at all to inflation would have undermined the trust that agents have in the central bank. Such a passive central bank that seemingly does not care about inflation would likely have triggered fears of endemic inflation and created expectations of future inflation. This in turn would have prevented the rate of inflation from declining at the same speed as the one observed. The interest rate hikes were meaningful in that they maintained low long-term inflation expectations.
This argument has some merit outside the realm of rational expectations. Agents with rational expectations, if they exist, must have realised that the inflation cycle was driven by an energy price cycle, and that the central bank was powerless to affect it. Clearly agents with rational expectations understanding this would not have raised their inflation forecasts when they observed a passive central bank not reacting to the inflation surge. The argument makes sense only if we assume that agents did not understand the true nature of the inflation cycle during 2022-24.
The cost of fighting inflation
Since the advent of quantitative easing (QE), the operating procedures of central banks have changed dramatically. From a reserve scarcity regime, central bankers moved to a reserve abundance regime. This came about as a result of quantitative easing, which led central banks to buy massive amounts of government bonds from banks and other financial institutions. The counterpart of these purchases was a dramatic increase of bank reserves (bank deposits held at the central bank). The result of this reserve abundance regime was that the money market interest rate got stuck at the zero lower bound. Then, when the central banks were forced to raise the interest rate to fight inflation in 2022, they could only do so by raising the rate of remuneration on bank reserves. This rate of remuneration became the new lower bound in the money market. As a result, fighting inflation using the rate of remuneration on bank reserves as the policy instrument, created collateral damage. With each rate hike central banks had to increase transfers of newly created monetary base to banks (see De Grauwe and Ji 2023).
In Table 1, we present these interest transfers to the banks made by the central banks of the Eurosystem during the years 2023 and 2024. One can see that they are sizable, amounting to between 0.8% and 0.9% of euro area GDP. Over the whole period that started with the interest rate hikes in September 2022 until today (January 2025) these transfers to banks have amounted to €270 billion. To give some perspective: the yearly spending by the EU amounts to €168 billion. This spending is the result of complex and often acrimonious decision-making processes and is accompanied by lots of conditionality for those at the receiving end of this spending. In contrast, the transfers to banks in the Eurosystem have been organised by stealth without any prior discussion, without being subject to a political decision-making process and with no strings attached.
Table 1 Yearly interest transfers to banks in the euro area (billion €)
Source: European Central Bank, ILM.W.U2.C.L020200.U2.EUR | ECB Data Portal
It can be useful to compare the interest transfers towards the banks with the profits made by these banks. We do this in Table 2 where we present the net profits of euro area banks during 2023-24. We can see that the interest transfers amounted to more than half of the banks’ net profits during these years. These are also two years when euro area banks made record-high profits. It should be stressed that the interest transfers had a one-to-one effect on banks’ profits. Banks did not have to use resources to earn these profits, as they must when they want to raise profits from lending. No credit and risk analysis had to be done on the deposit accounts they held at the central banks. With each increase of the deposit rate, profits increased without banks having to do anything, creating pure windfall profits. Table 2 therefore suggests that the interest rate transfers during 2023-24 had the effect of more than doubling banks’ profits in the euro area.
Table 2 Net profits of euro area banks and interest transfers as % of net profits
Source: European Central Bank
The counterpart of these large profits of banks was large losses incurred by the central banks of the euro area (for more detail, see Belhocine et al. 2023 and De Grauwe and Ji 2024). We show the losses incurred by the major central banks of the Eurosystem in 2024 in Table 3. The total losses of these central banks in 2024 amounted to almost €70 billion. As a result, these central banks stopped transferring yearly profits (seigniorage) to the national treasuries. As the losses of central banks are sizable, many of them have announced that for many years to come there will be no transfers to national treasuries.
Thus, it appears that the anti-inflationary policies of the ECB were extremely costly. The price paid for the fight against inflation amounts to several hundred billion euros. In the previous section we found that the contribution of the ECB in lowering inflation is minimal. It is, therefore, very likely that the cost-benefit ratio of the successive interest rate hikes by the ECB is very high. But there might be other costs besides those we have highlighted in this section.
Table 3 Losses incurred by major central banks in the Eurosystem
Source: National Central Banks accounts
Other costs of the fight against inflation
When economists think about the costs of fighting inflation, they have something else in mind than the costs we have been analysing here. These are the costs in lost output and employment. Given the existence of price rigidities, interest rate hikes will generally lead to a decline in both output and prices. Thus, the costs of the fight against inflation also include the lost output that was necessary to bring down inflation.
How large are these costs? In contrast with the interest transfers, the cost of output losses cannot be measured with the same precision. In Figure 4 we show the interest rate (quarterly) and the real growth of GDP in the euro area. When the ECB started to raise the interest rate in 2022, these interest rate hikes did not seem to have mattered much for the subsequent growth rates. In fact, we observe a slight increase in these growth rates from 2022Q4 throughout the period of interest rate hikes.
This may seem surprising, but it really is not, for two reasons. First, the sharp decline in energy prices from early 2023 (see Figure 3) acted as a positive supply shock. This had the effect of reducing inflation. It also led to an increase in real disposable income of consumers in the euro area. This increased aggregate demand and more than compensated whatever negative demand effect resulted from the interest rate hikes.
Second, as we have shown elsewhere (see De Grauwe and Ji 2024) the interest rate hikes lost much of their effectiveness by the operation of the ‘equity effect’. As the ECB raised the interest rate it also made increasingly large transfers to banks. These had the effect of more than doubling banks’ profits (see Table 2). With higher profits, banks’ equity improved giving them an incentive to lower credit standards and/or to increase the supply of loans (for empirical evidence see Fricke et al. 2023, De Grauwe and Ji 2024). In doing so, banks contributed to reducing the effectiveness of the restrictive monetary policies of the ECB. Put differently, the massive transfers of base money to banks during rate hikes deranged the transmission of monetary policies. The latter became less effective in reducing aggregate demand and inflation.
Thus, it appears that the ECB’s policy of monetary restriction did not have visible effects on GDP growth rates. It also follows that if there were output losses resulting from this monetary restriction, these losses are likely to have been small.
Figure 4 GDP growth and interest rates in the euro area
Source: Eurostat and ECB
Fighting inflation at a lower cost
The ECB is unlikely to have contributed significantly to lowering inflation during 2022-24. As a result, the high cost of the interest transfers to the banks is likely to have exceeded the benefits of the interest rate hikes during that period.
But even if we grant the benefit of the doubt to the ECB and assume that its interest rate policy contributed significantly to lowering inflation, it could have done so at a lower cost. The ECB could have lowered the cost by introducing an asset-based system of reserve requirements (Schobert and Yu 2014). This consists in computing minimum unremunerated reserves as a percent of total bank reserves. Thus, if bank A has total bank reserves of 100 and bank B of 200, the ECB could have told these banks that, say, 50% of these bank reserves are unremunerated.
Such a system is perfectly feasible. In fact, the ECB applied it during 2019-22 when the deposits at the central banks carried a small negative interest rate. Banks had to pay 0.5% on their deposit accounts at the central bank. Banks did not like this, and the ECB designed a tiered system to please the banks. In this tiered system part of the bank reserves were exempted from interest charges.
The advantage of such a two-tier system is that the operating procedure of central bankers can be kept unchanged. The central bank continues to use the interest rate on bank reserves (the deposit rate) as its monetary policy instrument. The banks continue to have the same incentive to hold excess reserves, as these continue to be remunerated at the margin. However, the transfer of central banks’ profits to commercial banks can be reduced significantly. Assuming that the central banks were to block 50% of the existing bank reserves in the form of non-remunerated reserves, the interest transfers could also have been halved, thereby significantly reducing the cost of the fight against inflation.
Conclusion
We argued that while most probably the interest rate policies of the ECB during 2022-24 had a trivial effect on inflation, these policies came at a high cost. The latter finds its origin in the new operating procedure of the central bank that consists in remunerating the large stock of outstanding bank reserves. With each interest rate hike the central banks of the Eurosystem were forced to transfer billions of monetary base to the banks. Our estimates are that during the period 2022-24 the central banks transferred €270 billion to the banks. As a result, the banks made massive windfall profits, and the central banks made massive losses. This also implies that these central banks stopped transferring profits (seigniorage) to the national treasuries and will continue to do so for many years to come to clean up their balance sheets.
We argued that this was not inevitable. The ECB could have used a tiered system for remuneration of bank reserves. This would have allowed for a significant reduction of the interest transfers to the banks, while keeping the central bank’s operating procedure unchanged. The ECB chose not to do so and therefore is responsible for an interest rate policy that was quite ineffective while costing taxpayers a lot of money. Only bankers were made happy.
Source: cepr.org
