Reading consumer spending and retail sales figures is much less easy than usual and it is more important than ever to distinguish between nominal and real when analyzing the figures.
Since the beginning of 2020, household consumption has been very seriously disrupted, due to the temporary closure of certain sectors (especially services) and supply difficulties for certain types of goods, then the acceleration of price. Reading consumer spending and retail sales figures is much less easy than usual and it is more important than ever to distinguish between nominal and real when analyzing the figures.
Impressive consumption in nominal, but much less in real life
In the United States, household consumption is particularly impressive when considered in nominal terms. Not only has it returned above its levels at the start of 2020, but it is even well above its pre-crisis trend. Retail sales (largely focused on consumption of goods) were 20% above the pre-crisis trend in May, which is very spectacular. Consumer spending (which includes consumption of goods but also that of services) was in April just over 5% above its pre-crisis trend.
But given the very strong acceleration of inflation over the last 9 months, it is absolutely necessary to follow the evolution of consumption in real terms. In fact, it is remarkable to note that real consumption fell sharply during the first phase of the Covid crisis (from March 2020 to March 2021), which had been marked by the closure of many sectors of the economy, before return almost perfectly to its pre-crisis trend since March/April 2021. In other words, the “quantities” consumed have returned from March 2021 to their pre-crisis trend.
There is therefore a particularly clear dichotomy between nominal consumption and real consumption, which can lead to errors of assessment. When we decompose the deviation in nominal consumption into a “price” effect and a “quantities” effect, it emerges that nominal consumption fell below the pre-crisis trend from March 2020 to almost March 2021. exclusively due to “quantity” effects (closed sectors, supply problems) then above the pre-crisis trend from March 2021 almost exclusively due to “price” effects. We can see that March 2021 corresponds both to a relaxation of health measures in the United States and to the adoption/implementation of Joe Biden’s American Rescue Plan (new mailing of checks to households in particular).
Very marked sectoral distinctions
As we have just seen, real consumption in the United States has returned to its pre-crisis trend, but this masks an unprecedented distortion in the structure of consumption, since real consumption of durable and non-durable goods is still above its pre-crisis trend, while that of services is still clearly below. This “great distortion” of consumption during the Covid crisis, in favor of goods over services, is quite unprecedented in these proportions when we consider long histories.
In recent months, we have seen a rebalancing of real consumption in favor of services and once morest goods. Real consumption of goods, while at high levels, has not grown at all in real terms for a year. Moreover, real retail sales are now even in sharp contraction over one year. For certain types of “premium” goods rather reserved for high-end customers (such as jewellery, perfumery or pleasure boats and planes), we note however that real consumption is not weakening, which might be linked to the fact that the richest 20% of households captured around two-thirds of the excess savings built up during the Covid crisis.
Unlike goods, the consumption of services is still in a phase of rapid expansion, especially as some segments are still in the phase of reopening. For example, we can think of the entertainment sector (amusement parks, sporting events), enrollment in sports clubs or the hotel industry.
One of the important points to stress here is that there is naturally a certain degree of price elasticity: the higher the prices, the lower the quantities consumed. Here, we can remember that it is mainly the prices of goods that have increased in recent quarters. While the price of durable goods has fallen each year between 1995 and 2020, it has increased very sharply in recent months (peak at +11.5% year-on-year in January). Services inflation, which had always been higher than durable goods inflation since 1976, is now significantly lower. This quite atypical configuration from a historical point of view (and even unique in its magnitude) partly explains why the real consumption of goods is currently marking much more time than the real consumption of services.
Moreover, this high inflation in the price of goods, by weighing on the quantities sold, is probably one of the elements behind the excess storage observed recently in American mass retailing.
The very sharp acceleration in inflation over the past 9 months makes it even less relevant than usual to confine oneself to nominal variables in the analysis of household consumption. In the United States, household consumption is very spectacular when we consider it in nominal terms but much less when we consider it in real terms. Real consumption of goods has marked time in recent months and is even flirting with contraction. For its part, the real consumption of services is still progressing rapidly, because the return to certain segments which were very penalized during the Covid crisis has not yet been fully realized, but also because inflation in services is much lower than for goods, which is very atypical historically.
