Labor Market Impacts of the Green Transition: Evidence from a Contraction in the Oil Industry with Elisabeth Isaksen and Maria Nareklishvili. CESifo Working Paper No. 12057 (Conditionally accepted at the Journal of the European Economic Association)
The transition to a low-carbon economy requires a contraction of fossil fuel sectors, raising questions about the labor market costs of reallocation. We study the 2014 oil price shock as a natural experiment to examine the contraction of Norway’s oil industry. Using matched employer–employee data, we estimate long-run effects on earnings and employment using two complementary approaches. A difference-in-differences design shows moderate losses for all oil workers, while an event study reveals substantially larger and more persistent losses among displaced workers—up to 10% in earnings and 5% in employment nine years after displacement, especially for those with lower educational attainment. Although few displaced workers transition into green jobs, they are equally likely to enter green and brown (non-oil) sectors when accounting for the size of each destination sector. Earnings losses are larger for those entering green jobs rather than brown (non-oil) jobs, but smaller than for those entering other sectors. Decomposition results indicate that differences in establishment wage premiums—rather than skill mismatch—explain most of the observed gaps.
Does the Salience of Climate-Related Risk Affect Asset Prices? (Revise and resubmit at the Journal of Financial and Quantitative Analysis) Available at SSRN
Using regression discontinuity and difference-in-differences designs to compare prices of California homes across a new wildfire risk zone boundary, I identify null effects in the immediate vicinity of the new risk zone boundary. In contrast, prices of homes newly assigned to a riskier zone drop by 5.7% further away from the new risk zone boundary. These findings suggest that households do not respond to the risk salience triggered by the new risk zone assignment, but instead update their risk beliefs in response to re-zoning in a way that is consistent with the actual wildfire risk. [Press coverage]
Road Pricing: Travel Behavior and Public Support with Alice Ciccone and Gøril Andreassen. CESifo Working Paper No. 11867 (R&R at the Journal of the European Economic Association)
We conduct a large-scale randomized controlled trial to examine the effects of time- and location-specific, distance-based road pricing on travel behavior and driving externalities. Using financial incentives and a smartphone app that automatically tracks participants' travel behavior across different modes, we find that road pricing reduces driving externalities by 5.3%, implying a price elasticity of -0.07 to -0.15 for the external costs of driving. Our findings suggest that drivers of battery-electric vehicles (BEVs) are much less responsive to road pricing than drivers of non-BEVs. Furthermore, we find that providing information on the expected benefits of road pricing enhances public support for such policies, whereas experience with road pricing has little impact.
Cloudy Judgments: Weather, Behavioral Bias, and Home Prices with Andreas E. Eriksen. Housing Lab Working Paper No. 2025-2
Economic decisions involving durable goods often require households to assess value and long-term utility across future states of the world. While many studies document that such decisions can be shaped by behavioral biases, less is known about when in the decision-making process individuals are most vulnerable. Using high-frequency data from the housing market, we investigate how the timing of transient environmental factors influences expectations about future utility. We find that cloud cover on the days of open houses significantly reduces sales prices, while it has no measurable impact during the purchase stage. Furthermore, we find no evidence that cloud cover influences buyer interest, such as attendance at showings or the number of offers submitted. The results are consistent with an attribute salience mechanism, in which temporary environmental cues like sunlight shape what buyers notice or attend to when forming their initial valuation. While mood effects may contribute, we find little evidence of projection bias, suggesting that perceptual biases, rather than forecast errors, drive the observed pricing effects. The findings underscore the importance of early-stage perception in asset valuation and suggest that behavioral distortions can arise even when incentives for rationality are strong.
Sticky Asset Prices with Martin Schmalz and Dag Einar Sommervoll. Available at SSRN
We provide causal evidence of downward price stickiness in asset markets by exploiting an unexpected and geographically confined increase in residential electricity prices in Norway. Using comprehensive housing transaction data and a difference-in-differences design, we find that home prices in treated areas decline by 4.2% to 6.6%, while transaction volumes fall by 5% to 9%, i.e., a pattern inconsistent with frictionless price adjustment. The volume decline is driven by reduced listings, longer time-on-market, and increased unsold inventory. To identify the mechanism, we examine heterogeneous treatment effects: a frictionless or pure-uncertainty account predicts uniform responses, whereas mortgage lock-in—a rational response to a binding down-payment constraint—and loss aversion—a behavioral friction—predict concentration among specific sellers. Volume declines are concentrated among homeowners who are expected to incur nominal losses or to be locked in by low home equity, while homes with ample equity or nominal gains show no significant decline. Our findings provide evidence that nominal loss aversion and negative home equity affect asset price dynamics. These results have implications for macroeconomic models of housing markets, the transmission of monetary policy through wealth effects, and the design of policies to maintain market liquidity during downturns.
Panic Investment in Durable Goods: Evidence from Solar Panel Adoption in Norway with Massimo Filippini
This paper examines how projection bias—the tendency to over-extrapolate current conditions into the future—shapes household investment in durable goods. Using municipality-level data and a sharp regional electricity price divergence within Norway during the 2022–2023 European energy crisis, we estimate the causal effect of the price shock on residential rooftop solar panel adoption. Municipalities exposed to high electricity prices saw a 556% yearly average increase in new solar capacity and a 458% yearly increase in solar installations relative to unexposed municipalities. Second, we estimate the projection bias using an approach combining reduced-form causal identification with a parsimonious structural model to interpret quasi-experimental estimates. We find that households put about 70% weight on current prices when evaluating long-run profitability. Our findings provide suggestive evidence that projection bias can influence rational investment criteria, with implications for microeconomic theory and policy.