We use transaction-level card expenditures to estimate the effect of Germany’s temporary value-added tax (VAT) in 2020. We use card expenditures in Austria as control because no analogous VAT cut was implemented in Austria and spending of Austrian households provides a stable benchmark to compare spending of German households against. As predicted by consumption theory, we find the VAT cut in-creased consumption growth more for durable goods, with a stronger effect close to the end of the cut. The annualized growth rate of expenditures for durables increased by 7.5 percentage points (pp), with a particularly strong increase of up to 23.6 pp for consumer electronics. The consumption growth rate for semi-durables increased by 5.6 pp, and for non-durables by 2.6 pp. The implied effect on durable spend-ing is four times smaller compared to survey-based estimates in Bachmann et al. (2023), and a fiscal shortfall that is 23 bn larger.

Changes to loss offset rules are often used as a fiscal stimulus measure in times of crisis to alleviate firms’ constraints and support the economy. The effect of these measures has been assessed in the literature using large listed firms, but there is no evidence on the effect of these policies on small private firms. I study a temporary change in the loss carry-back period implemented in the Netherlands over 2009-2011 using administrative and tax return data on small private corporations. Using a difference-indifference set up and matching techniques, I show that an additional year of carry-back has a significant effect on treated firms’ investments, but no significant effect on firms’ survival, employment and profits. The effect on investments is not driven by equity injections but rather by debt increases. Crucially, the significance of the effect on investments is conditional on the

size of the additional carry-back.

Using quarterly data on UK government spending forecasts alongside the household surveys dating back to 1977, we provide new evidence on heterogeneous effects of government spending policies on household consumption. Specifically, while at an aggregate level the on-impact consumption multiplier is about 0.5 and the effect fades only gradually, the consumption effects of spending shocks are substantially greater for (1) households with mortgage debt and (2) households in the left tail of the income distribution. As is well known, mortgagors, unlike outright homeowners, tend to have large illiquid assets, but little liquid wealth. Our results thus suggest that household balance sheets, together with their income levels, play a key role in the transmission of government spending policies in the UK.