In response to the Covid-19 pandemic, many governments around the world used direct transfers to households to generate fiscal stimulus. Similar policies had already been used in the 2001 recession and during the Global Crisis, where they proved to be effective in stimulating household spending (Johnson et al. 2006, Parker et al. 2013). However, the recession caused by the Covid-19 pandemic differs substantially from prior economic downturns. Restrictions imposed by governments to slow the spread of the virus and the fear of infection play a role for consumption decisions. Therefore, it is an open question whether direct transfers work as well as they appeared to have done in prior recessions. For example, Auerbach et al. (2021) show that multipliers from defence spending were positive only in cities that were not subject to stay-at-home orders. Estimates of the marginal propensity to consume for the US stimulus checks from the Coronavirus Aid, Relief, and Economic Security (CARES) Act vary substantially (Baker et al. 2020, Coibion et al. 2020, Parker et al. 2022).
A direct transfer of €450 per child
The German federal government also enacted a direct transfer payment to households with children, the so-called ‘child bonus’. The intention of this policy was to cushion the strains Covid-19 restrictions placed on families as well as to bolster aggregate demand. We confirm, using survey data, that households with dependent children report higher income losses in the beginning of the Covid-19 pandemic. In a new study, we investigate whether the child bonus did indeed stimulate consumption to a measurable extent and what factors influence its effectiveness (Goldfayn-Frank et al. 2022). Our study utilises scanner data from the market research institution Gesellschaft für Konsumforschung (GfK), which records the daily consumption expenditure of almost 10,000 households on non-durable goods such as food items, and semi-durable goods such as clothing. We combine the daily data on household expenditure and the randomly distributed payment dates of the child bonus to identify its effect on spending. Thus, we compare the expenditures of two households that differ only in that one has already received the child bonus while the other has not. One advantage of our study is that we observe actual spending behaviour by households and do not have to rely on surveys.
To give us a first broad sense of whether the child bonus increased consumption spending, in Figure 1 we compare the average monthly consumption expenditure of households with children with that of households without children between July 2020 and June 2021.
Figure 1 Monthly spending by households with and without children
The bonus was paid out in three tranches: €200 per child in September 2020, €100 per child in October 2020, and €150 per child in May 2021. Looking at Figure 1, we can see a month-on-month increase in spending by households with children in September 2020. In the case of households without children, however, average expenditure remained constant across the two months in question. Things are different when it comes to the second and third payment. In October 2020, spending by both groups of households runs along parallel trajectories. In May 2021, households with children even spent a little less on consumption than they had done in April, while households without children spent around the same amount. These descriptive patterns suggest that only the first tranche of the child bonus led to increased spending.
To assess the impact of the child bonus on household expenditure more systematically, we perform a difference-in-differences analysis comparing households that have received the transfer to those who have not received it (yet).1 We express our results as an estimate of the marginal propensity to consume, that is to say the percentage of the transfer payment spent within the month.
Figure 2 Estimates of daily effects on total spending
Only the first transfer increased spending
Our analysis shows that the child bonus had a relatively small effect on household spending. For the first payment, we estimate a marginal propensity to consume of about 12%. In other words, out of €1 of child bonus, households spent about 12 cent in the month in which the transfer was received. As Figure 2 shows, we observe parallel spending trends in the days before the households receive the transfer. In the days after the receipt, the treated households increase their total spending temporarily. The effect was concentrated in the non-durable goods category and was driven by households in districts with lower Covid-19 case rates. Households with a low income or liquidity constraints also exhibited a stronger response, though only a small share of households report such constraints. In contrast, households with higher saving rates responded only weakly to the child bonus. The spending effect is not systematically related to the local labour market situation or the stringency of local coronavirus response measures. Furthermore, we do not find that there was any consumption effect caused by the announcement of the transfer. The number of contacts of households due to economic activity, measured by the number of shops they visited, rose due to the child bonus. Online shopping played a relatively minor role. This implies that there could also be a feedback effect from the increased economic activity to infection rates. Such a mechanism is a feature of models that integrate both macroeconomic and epidemiological dynamics (Eichenbaum et al. 2021).
At the time of the second and third transfer, infection rates were a lot higher
We do not identify any significant spending effect for the second and third instalment of the child bonus. Taken together, this yields an aggregate marginal propensity to consume of just 5% for all three tranches of the child bonus added together. The absence of a consumption response for the second and third transfer payments could have to do with continued high saving rates among the population as well as substantially higher infection rates during the months of the second and third payments. We do not find evidence that the different transfer sizes play a role in explaining the null effect for the latter payments. Finally, even if spending on durable consumer goods and services, which is not covered by our data, was to exhibit a similar increase, the marginal propensity to consume would still be only about 14%.
Redistribution rather than stimulus
Overall, the child bonus appears to have exerted only a limited consumption stimulus. This is partly because of the specific nature of the pandemic context, which acts to inhibit the transfer’s effectiveness. Our result is consistent with the findings of Parker et al. (2022), who report a marginal propensity to consume of around 10% for the Economic Impact Payments, which were also paid out as direct transfers to US citizens in 2020. By contrast, other research has shown that the temporary VAT cut in Germany provided an effective boost to consumption (Bachmann et al. 2021). The child bonus therefore served not so much as a stabiliser of economic activity but rather acted as a redistributive instrument.
Authors’ note: The views expressed here do not necessarily reflect the opinion of the Deutsche Bundesbank or the Eurosystem.
Auerbach, A, Y Gorodnichenko, P B McCrory and Daniel Murphy (2021), “What Covid-19 teaches us about fiscal multipliers”, VoxEU.org, 23 December.
Baker, S, R A Farrokhnia, M Pagel, S Meyer and C Yannelis (2020), “Income, liquidity, and the consumption response to the COVID-19 pandemic and economic stimulus payments”, VoxEU.org, 17 June.
Bachmann, R, B Born, O Goldfayn-Frank, G Kocharkov, R Luetticke and M Weber (2021), “A Temporary VAT Cut as Unconventional Fiscal Policy”, VoxEU.org, 20 November.
Coibion, O, Y Gorodnichenko and M Weber (2020), “How US consumers use their stimulus payments”, VoxEU.org, 8 September
Eichenbaum, M, S Rebelo and M Trabandt (2021), “The macroeconomics of epidemics”, The Review of Financial Studies 34(11): 5149–5187.
Goldfayn-Frank, O, V Lewis and N Wehrhöfer (2022), “Spending effects of child-related fiscal transfers”, CEPR Discussion Paper 17058.
Johnson, D S, J A Parker and N S Souleles (2006), “Household expenditure and the income tax rebates of 2001”, American Economic Review 96(5): 1589-1610.
Parker, J A, N S Souleles, D S Johnson and R McClelland (2013), “Consumer Spending and the Economic Stimulus Payments of 2008”, American Economic Review 103(6): 2530-2553.
Parker, J A, J Schild, L Erhard and D Johnson (2022), “Household Spending Responses to the Economic Impact Payments of 2020: Evidence from the Consumer Expenditure Survey”, NBER Working Paper 29648.
Sun, L and S Abraham (2021), “Estimating dynamic treatment effects in event studies with heterogeneous treatment effects”, Journal of Econometrics 225(2): 175-199.
1 We use the estimator proposed by Sun & Abraham (2021) which is robust to heterogeneous treatment effects.