RaboResearch - Economic Research

Inflation: Ambient or transient?


  • Markets have been in the grips of reflation, although the rise in market-based inflation expectations is not being echoed so far by households
  • A jump in Eurozone headline inflation for January has played into these reflation fears, although we show that this is mostly due to temporary and cost-push factors, including Covid-19 related basket changes by Eurostat
  • Cost-push inflation is likely to be transient and it could even turn into disinflation. Indeed, without durable government support beyond covid-19 and without a change in wage formation, demand-led inflation is likely to remain absent
  • We believe the ECB will ultimately look through this temporary volatility in inflation. After having been wrong (i.e. over-estimating) on inflation for most part of the recent decade, the ECB cannot afford to make any policy mistakes


Figure 1: Rising inflation break-evens, being followed by higher long-term rate expectations
Figure 1: Rising inflation break-evens, being followed by higher long-term rate expectationsSource: Bloomberg, Macrobond

In recent weeks, financial markets have been in the grips of reflation. Inflation break-even rates have reached their highest levels since the pandemic, led by an even stronger rise in the US. Long-term interest rate expectations have suddenly started to rise this year and while inflation curves broadly flattened on the back of higher oil prices since April 2020 –the markets’ usual reaction to a rise in oil prices–, in recent weeks we have seen some steepening of the inflation curve. At the same time, January’s HICP release made for a significantly positive surprise.

A confluence of factors has contributed to the development highlighted above, but the million dollar question is whether these factors are fundamental enough to mark this as ‘sustainable’. To be sure, there are clearly arguments supporting the reflation narrative –see the thought piece that we wrote in June 2020 (will inflation become a monetary phenomenon again?). But some key pieces of the inflation puzzle –primarily wage growth– are still missing. To our minds, recent price pressures remain of the cost-push variant, which are less durable in the worst case or could even be long-term disinflationary in the worst case.

Households’ expectations still muted

Figure 2: Inflation perceptions households
Figure 2: Inflation perceptions householdsSource: Macrobond

For starters, the rise in market-based inflation expectations is not being echoed so far by households (see figure 2), although businesses have been reporting higher selling price expectations since April 2020, albeit still from relatively subdued levels.

This seems closely related to lockdowns and subsequent reopening of parts of the economy as well as to accelerating price pressures in the foreign trade complex, e.g. costs related to shipping. To be more specific, import and producer prices of intermediate goods have started to rise in recent months, but they are still down in y/y terms and this hasn’t spilled over materially yet to consumer and capital goods. These price pressures in the pipeline therefore very much align with the story unfolding on global commodity markets, where we are witnessing a sharp rise in energy and non-energy (particularly food) prices, as well as with supply-chain issues in certain sectors.

What has driven the Eurozone HICP in recent months?

The sharp rise in headline inflation for January (from -0.3% y/y to +0.9% y/y), as was first reported by Eurostat on the 3rd of February, served as a wake-up call for markets; since then, long-term swap rates have risen by some 20bp. So let’s drill down into what caused this sharp rise in Eurozone inflation. We can show that the majority of the rise in headline inflation comes down to four specific factors, although this analysis still leaves 0.2-0.3%-points unexplained.[1] 

Figure 3: Headline inflation before and after…
Figure 3: Headline inflation before and after…Source: Macrobond, RaboResearch

First of all, there was considerable impact from indirect taxes (both on energy as well as a broad range of items through the VAT hike in Germany). In total, this impact was around +0.53%-points on the y/y rate. Including government-administered prices, this even amounts to 0.6%-points. There was also a considerable contribution from higher energy prices, which we estimate at around 0.18%-points. However, on top of these (temporary) factors, Eurostat made considerable changes to the weights of the HICP components in the basket, to ‘better’ reflect the Covid-19-affected consumption patterns during 2021. The appendix at the end of this piece gives an overview of most notable changes in those weights.

In terms of the overall contribution of this effect, we estimate this to have been around +0.27%.[2] Figure 3 gives an indication of what inflation would have been without these basket changes. The detailed exercise highlights, however, that there was one single component that was responsible for no less than a 0.15% m/m contribution, which was ‘package international holidays’, where prices fell 18.7% m/m in January after rising 15.1% m/m in December. In other words, basket changes amplified the seasonal effect in that category.

Figure 4: ‘Quadrant’ analysis of most notable changes in inflation and/or weights for January
Figure 4: ‘Quadrant’ analysis of most notable changes in inflation and/or weights for JanuarySource: Macrobond, RaboResearch

In figure 4 we show a ‘quadrant’ analysis, where the most notable changes in contribution for 282 HICP items (at the 5-digit level) are plotted on two axes: the change in the basket weight on the vertical axis and the change in the year-on-year rate (between December 2020 and January 2021) on the horizontal axis. In each quadrant we added the top items, which provides some interesting insights into what were the key drivers behind the uptick in inflation in January and whether or not this was amplified or mitigated by changes in the weights. The most interesting ones are in the upper-right and lower-left quadrants as there both forces strengthened each other (i.e. higher inflation and higher weight, versus lower inflation and lower basket weight).

Broadly speaking we can see that inflation got a boost in January from higher energy costs, fresh vegetables, household furniture, tech gadgets and ticket prices; but in the case of package holidays, tickets, garments and petrol this was attenuated by a decline in their weight, whereas the impact was amplified for electricity, fresh veggies, household furniture etc. On the flip side, there was a notable decline in inflationary pressures in holiday/recreation services and restaurants/café’s as well as in house rentals, fresh fish/fruit, certain meats and hospital services. Moreover, particularly the former lost weight in the inflation basket, lessening their drag on headline inflation.

Intuitively, these new HICP weights make perfect sense, as it reflects the shift of consumption from shuttered services to goods. However, in the event, it did clearly result in a higher HICP reading for January than would have otherwise been the case. Now, of course, the key question is what will happen when the situation returns ‘back to normal’ – both because consumption may be spread over more items again, and because the basket may be readjusted next year. Would that mean that we could see a reversal of this development? For the moment, we would argue that the answer to this question hinges on whether demand would fully return to pre-Covid levels or not. In our view it is actually more plausible that it will not, as a gradual withdrawal of government support is likely to expose underlying weaknesses and unleash an overdue wave of bankruptcies and unemployment, which will prevent a quick ‘v-shape’ recovery in demand. In this case, it is likely that we will see a reversal in consumption patterns (at least partially) and as such a negative composition effect on inflation of up to 0.3%-points as early as next year.

Governments and wages hold the key(s)

Figure 5: A clear break with the GFC and sovereign debt crisis
Figure 5: A clear break with the GFC and sovereign debt crisisSource: RaboResearch, Macrobond

As we recalled in the introduction, there are clearly some arguments that support the reflation narrative. In particular the notion that central banks and governments are now coordinating and working jointly to support demand – in the face of Covid-19 supply restrictions. Not earlier in recent history have we seen such a strong joint rise in money supply growth as well as government deficits, as illustrated in figure 5. As such this is a clear break with the response to the GFC in 2008-09 as well as the European sovereign debt crisis in 2011-13, where either monetary policy or fiscal policy lacked oomph.

A key question, however, is whether governments will maintain their budgetary support long enough so as to prevent demand from weakening in the post-Covid recovery. On this front, the jury is still out. On the one hand politicians in the likes of Germany may be preparing themselves for a return to ‘austerity’, On the other, the debate at the EU-level has shifted a lot in recent years, and a swift return to the old ‘stability and growth pact’ world does not seem very likely.

Rising costs due to Covid-19 related supply restrictions and/or geopolitical shifts can also be counted as upward forces to inflation, although – seen in isolation – this would still very much be a temporary push rather than a permanent factor. Arguably, the rising global shipping costs, the rise in global commodity prices and the HICP basket re-weighting all fall into this same category.

But the missing link (or missing piece of the puzzle) amidst all these deliberations is the labour market and in particular the future development of wages. Although unemployment has been coming down in recent months, it has stayed well above the ‘NAIRU’ and this is likely to remain the case in the foreseeable future. Of graver concern is whether people who have withdrawn from the labour market altogether (as seen in the activity rate) will be able to return soon after Covid-19 restrictions are lifted. At the very least it would seem that a full normalization of the labour market is not to be expected before mid-2022, in line with our GDP forecasts.

Figure 6: Not pinning our hopes on ‘wages’ just yet
Figure 6: Not pinning our hopes on ‘wages’ just yetSource: Macrobond

This, in turn, is likely to keep a lid on wage growth – as is already visible in the development of negotiated wages and labour costs in in recent quarters (figure 6). Hence, even under sunnier conditions in 2021-2022 the labour market is unlikely to exert much upward pressure on wages. The wage-price spiral that has been effectively broken in Europe in the course of the 1980s, alongside a retreat of labour unions is not something that can be reinstated or repaired at short notice. The story of ‘reluctant wages’ is something that is known all too well in Japan and the situation in Europe does not seem to be very different.

Although there is a significant monetary overhang – which may ultimately feed into inflation – and geopolitical/deglobalization forces could be another inflationary factor in the long-term, we believe that two key pre-requisites for these inflationary pressures to become more sustained in the medium-term are still missing: a sustained rise in domestic wage pressures[3] and continued budgetary stimulus. The latter, we do acknowledge, is certainly more of a wildcard at the moment. If governments – with EU backing – will continue their budgetary stimulus beyond the Covid-crisis, this may actually turn into a more favorable factor. For now, however, we would argue that cost pressures should prove to be transitory at best or even disinflationary in the longer-term.

Outlook: ‘wrongflation’ then?

Against this backdrop and assuming conservative projections for exchange rates and global commodity prices (steady levels), we can see clearly that base effects and recent upward pressure on global energy prices are likely to push the headline inflation rate higher in coming months. The projected inflation rate and its key contributors is shown in figure 7. Although slowing food price inflation is expected to offset some of these upward pressures initially, it will join forces with energy from the second half of 2021 onwards. The pass-through of global food prices (+17.4% y/y in January) tends to be slower and has been mitigated by a strengthening euro, but will still be felt later this year.

Core inflation (excluding energy and food) is projected to be stable in the coming months, but to tick higher again around the middle of 2021 due to base effects. However, as we enter 2022, not only those base effects go into reverse (including that of the VAT hike), but underlying forces should also lead core inflation back below the 1% threshold. It is towards the end of 2022 that we may see a slow and very gradual recovery in underlying inflationary pressures. But it will not be before the end of 2023 that inflation has returned to levels which the ECB may see as broadly in line with its inflation target.

Figure 7: YoY inflation projections – contributions of key components
Figure 7: YoY inflation projections – contributions of key componentsSource: RaboResearch
Figure 8: Inflation projections annual
Figure 8: Inflation projections annualSource: RaboResearch

All in all, inflation is likely to rise considerably in the months ahead, at some point even pushing it temporarily above the 2%-mark. But, as explained, most of the forces underpinning this development in 2021 are of the cost-push type rather than demand-pull. The rise in headline inflation may be taken by ECB hawks as a sign that inflation risks are on the rise, but for now we believe the ECB will look through this temporary volatility in inflation. After having been wrong (i.e. over-estimating) on inflation for most part of the recent decade, the ECB cannot afford to make any policy mistakes. That also means that we expect it to use all of its instruments to suppress inflation-induced rate pressures if being felt ‘unwarranted’.

Appendix – basket weight changes at 3-digit level (ordered from low to high)

Table 1: HICP items at the 3-digit level, weights and changes > 0.05%
Table 1: HICP items at the 3-digit level, weights and changes > 0.05%Source: Macrobond, RaboResearch


[1] This results from taking the difference between the m/m inflation rate using the official index (which uses the old weights for December and new weights for January) and the hypothetical m/m inflation rate using our own calculations of a January HICP if the basket weights had not changed.

[2] This is of course hypothetical: what would inflation have been had the basket in 2020 been similar to that of 2021. 

[3] Of note, in that respect, is the recent discussion on minimum wages in several countries (such as Italy). Even at EU level there is talk of introducing ‘fair minimum wages’ with the Commission having launched a second-stage consultation with social partners last year. But for now these talks seem to be in an early stage and not supported by all member states.

Elwin de Groot
RaboResearch Global Economics & Markets Rabobank KEO
+31 6 1389 2916

naar boven