Expectation Formation in a New Environment: Evidence from the German Reunification , with Johannes Wohlfart. Journal of Monetary Economics, August 2019 (first version: September 2017)
Unconventional Fiscal Policy at Work, with Rüdiger Bachmann, Benjamin Born, Georgi Kocharkov, Ralph Luetticke and Michael Weber. AEA Papers and Proceedings, 2023 vol. 113, pp. 61-64
A Temporary VAT Cut as Unconventional Fiscal Policy, with Rüdiger Bachmann, Benjamin Born, Georgi Kocharkov, Ralph Luetticke, Michael Weber. Review of Economic Studies, revise & resubmit. NBER working paper, 2023
Abstract: We exploit the temporary VAT cut by three percentage points in Germany in the second half of 2020 to study the spending response to unconventional fiscal policy. We use survey and scanner data on households’ consumption expenditures and their perceived pass-through of the tax change into prices to quantify the effects of this VAT policy. The temporary VAT cut led to a relative increase in durable spending of 37 percent for individuals with high perceived pass-through. Semi-durable spending also increased. According to a back-of-the-envelope calculation, the VAT policy increased aggregate consumption spending by 26 billion Euros, or 1.6 percent.
A Choice-Based Approach to the Measurement of Inflation Expectations, with Pascal Kieren and Stefan Trautmann. Journal of Monetary Economics, reject and resubmit. April 2023
Abstract: In applied research and policy analysis, economists have increasingly relied on measures of inflation expectations and uncertainty elicited via density forecasts. This method, where respondents assign probabilities to pre-specified ranges, has been subjected to criticism particularly in the recent times of high and volatile inflation. We propose a new method to elicit inflation expectations, which is rooted in decision theory and can be implemented in a standard survey. We demonstrate that it leads to well-defined expectations with central tendencies close to the corresponding point forecasts and to lower forecast uncertainty than density forecasts. The reported difficulty of the questions and survey time required are less or similar to the current standard. In contrast to the density forecasts, our method is robust to differences in the state of the economy and thus allows comparisons across time and across countries. The method is portable in the sense that it can be applied to different macroeconomic measures.
Spending effect of fiscal transfers in a pandemic, with Vivien Lewis and Nils Wehrhöfer. CEPR Discussion Paper, 2022
Abstract: As part of Germany's fiscal response to the Covid-19 pandemic, parents received three payments totalling 450 euros per child. Randomization in the payment dates and daily scanner data allow us to identify the effects of these transfers on household spending. We find a significant but small spending effect of the first transfer, with an estimated marginal propensity to consume of about 12%. The effect is higher for low-income and liquidity-constrained households, and in areas with lower infection rates. The second and third payment failed to increase spending. Our results indicate that the child bonus was redistributive rather than stimulative.
Abstract: Do recessions harm investment in technology and thus future aggregate supply? We provide novel evidence on this question using unique, granular data on innovation investment in R&D and diffusion from a representative survey of German firms. Our data allows to identify the crisis-induced innovation investment cuts with mean conditional reductions of -65% (R&D) and -70% (diffusion) relative to pre-crisis investment plans, concentrated in 20% and 25% of firms respectively. We estimate that a 1% cyclical output drop translates into a -0.3% fall in innovation investment. Firm-level financial constraints amplify the innovation reductions. Our findings suggest that short-term shocks affect aggregate supply over at least the medium term, challenging the exogenous technology assumption and the resulting dichotomy between business cycles and long-run growth in standard models of aggregate fluctuations. We show that demand shocks are among the main causes of the cyclical technology investment cuts, supporting the view that demand shocks can manifest as technology shocks. We formalize our micro-level results in a New Keynesian model with endogenous growth through investment in R&D and technological diffusion which determines cycle and trend jointly in general equilibrium.
Beyond the Short Run: Monetary Policy and Innovation Investment, with Michaela Elfsbacka-Schmöller and Tobias Schmidt.
Using a novel approach implemented in a unique survey of German firms, we study how monetary policy affects firms' innovation investment. We find that the ECB's policy rate hike of cumulative 450 basis points at the end of 2023 led 33% of firms to cut innovation spending by 67% on average, with effects persisting through 2024-2025. We show further that while rate hikes discourage innovation investment, cuts stimulate firms' innovation spending. We further demonstrate that forward guidance communication can provide significant additional stimulus, suggesting longer-term, supply-side effects of forward guidance. Monetary policy transmission operates through both financial and demand channels, with stronger effects for smaller firms, those with a higher share of bank loans, and those with weak expectations about demand. Our findings indicate persistent effects of monetary policy on longer-term aggregate supply, challenge long-run monetary neutrality and suggest policy-endogeneity of the natural rate of interest R*.
Gender Differences in the Marginal Propensities to Consume and Repay Debt, with Louiza Bartzoka and Nathanael Vellekoop.
Abstract: Do women respond to income shocks differently than men? We use questions from the NY Fed Survey of Consumer Expectations to study marginal propensities to consume and pay debt. We nd that women are more inclined to repay debt after an unexpected increase in income, where men have a larger marginal propensity to consume. After a decrease in income, both men and women state large reductions in spending, but women would reduce spending more. These differences in responses can be explained by differences in the debt-to-income ratio, the degree of discouraged borrowing, and liquidity constraints. Together these three variables explain about 40% of the gender differences in the marginal propensity to repay debt, and 75% of differences in the marginal propensity to borrow. A simple model with occasionally binding borrowing constraints can rationalize the gender differences in the marginal propensities to consume and repay debt.
Cross-Country Differences in Household Financial Decisions: A Structural Approach with Survey-Based Expectations, with Louiza Bartzoka and Georgi Kocharkov. Revised version in progress.
Abstract: We examine the effect of subjective expectations on household savings decisions, by developing a model of savings with incomplete markets. In a departure from the literature, we directly incorporate rich data on subjective household expectations from the ECB Consumer Expectation Survey, without imposing additional structure on belief formation. This allows us to leverage the benefits of empirical data, while being able to run counterfactual experiments through our calibrated model. Our analysis focuses on six large EU economies, and on the period of September 2020 to March 2022, during which households across Europe had negative expectation errors regarding future economic growth and expected higher level of real rates. Our approach gives us the ability to focus on how expectations interact with households' income level and implied liquidity constraints, as well as with their preferences. It hence aids our understanding of whether and how subjective expectations influence households' savings decisions, a topic at the core of the discussion on the role of forward guidance in economic policy.
Personality traits and financial decisions of the households, November 2018 (first version: December 2015)
Abstract: Who makes financial decisions in a household and how do the characteristics of the decision-maker affect subsequent financial choices? I use the Health and Retirement Study to directly identify the CFO of the family and consider the effects of personality traits on financial decisions above and beyond cognitive abilities, risk aversion, time preference and economic factors. Several findings are reported. First, I show that cognitive as well as non-cognitive abilities of both spouses matter for taking responsibility over financial decisions in the household. Second, I document an age-related shift: for elder households, non-cognitive abilities matter substantially more than for younger households. This difference becomes particular pronounced in risky financial choices: while for elder households, personality traits are among the most significant predictors of stock ownership as well as higher share of wealth invested in stocks, for younger households the effects of personality traits are negligible.
Borrowing from Family and Friends: Preferences, Peers and Personality , with Nathanael Vellekoop. October 2018
Abstract: We study determinants of the demand and supply of loans from family and friends using two Dutch household surveys. Despite having one of the most developed consumer loan market, the most common source of noncollaterilized debt of households is family loans. In line with theories of gift exchange and costly favors, we find that social preferences are a driving factor for both informal lending and borrowing. General trust matters for borrowing, but not for lending. Households who perceive that their social circle earn more are more likely to borrow from family and friends. We also document that non-cognitive traits (notably conscientiousness and self-esteem) are important determinants of the supply and demand of loans from family and friends. Family loans seem to be part of a social relationship, and are more complements than substitutes to formal consumer loan arrangements.
Intra-generational transfers and risk sharing in times of crisis, with Jian Li
Abstract: We use panel data from the Health and Retirement Study to answer the questions, if crises influences the transfers among parents and children via difference-in-difference approach. For identification, we use two facts: (1) while .com crises has affected mostly stockholders with large portion of economy remaining unaffected, the Great Recession affected the whole economy, (2) only a share of retired households has investments in stocks, and therefore were affected by the internet bubble crises. Therefore, we are able to consider two groups of parents (stockholders vs. non-stockholders) and two crises events. Thus, we are able to explore the responses of the children to wealth shocks to their retired parents during the .com crises, as well as response of retired parents to the income/wealth shocks to the working children during the great recession.
Mental Health and Financial Decisions of Households, with Yuri Pettinichi