Capital-Labor Substitution Using a State-Space Model
(With Anders F. Kronborg and Peter P. Stephensen)
(R&R Oxford Bulletin of Economics and Statistics)
This paper presents a new state-space framework for simultaneously estimating the capital-labor elasticity of substitution and recovery of the unobserved factor-augmenting technical change variable. We specify technical change as a non-linear process that captures a long-run trend and potentially several persistent fluctuations. In a simulation study, this approach outperforms both the linear and Box-Cox specifications of the technology process. In an empirical exercise containing 16 OECD countries, our framework delivers the desired properties, such as an elasticity below unity and several persistent fluctuations in technical change, particularly in the 90s and after the financial crisis.
[Working Paper] [Slides] [Codes]
Firm Heterogeneity and the Transmission of Foreign Supply Shocks [NEW]
(With Nicolai Waldstrøm)
In this paper, we investigate the role of within-sector heterogeneity in the transmission of foreign supply shocks. We start by establishing causal evidence of how a 10% increase in import prices leads to a 7% decline in real output and a 2.5% increase in prices. We then develop a partial equilibrium model that captures the heterogeneity within sectors, with firms varying across five dimensions: productivity, materials share, import share, export share, and IO-linkages. Our findings reveal that within-sector heterogeneity amplifies the response to foreign shocks by 32% in partial equilibrium. This amplification occurs primarily because larger firms also import the most. However, when considering the broader implications in general equilibrium, a counter-intuitive result emerges. Within-sector heterogeneity dampens the impact of foreign shocks by 11% in this setting. This surprising effect is attributed to the fact that larger firms, while import-intensive, are less intertwined with the domestic economy. In summary, our study underscores the complex interplay of within-sector heterogeneity in the transmission of foreign supply shocks, demonstrating its amplifying effect in partial equilibrium and its contrasting dampening effect in general equilibrium.
Complementarity vs. Substitutability of
Imports and Implications for the Comovement of Business Cycles
(Work in progress)
This paper employs Danish firm-level data to investigate the role of the firm's importing decisions in driving the comovement of business cycles. First, I find that firms with direct and indirect trade linkages have a significantly higher correlation with the growth rate of foreign GDP. At the aggregate level, removing these trade linkages would imply a decrease in the correlation of business cycles by 27%. Second, I find that the comovement of business cycles is mainly driven by the import of goods distantly related to the goods that the firm produces, instead of goods closely related. This suggests that goods from the same industry are more substitutable to the firm's own production than distant industries. I next want to analyze how this difference in substitutability affects the comovement of business cycles in economic models.
Policy papers
Estimating trade elasticities for Denmark
(With Tamás Vasi and Christian Vikkelsø )
(DREAM working paper, 2023)
We estimate the elasticity of substitution between domestically produced goods and imported goods. In particular, we estimate the homeforeign elasticity of substitution - also known as the macroelasticity. We use a methodological approach based on Feenstra et al. (2018), in which we distinguish the elasticity of substitution of imported goods from different countries of origin (microelasticity) and the macroelasticity. The median microelasticity is 2.97, while we estimate the macroelasticity to be 1.81.
Estimating trade elasticities in MAKRO
(With Anders F. Kronborg and Kristina A. Poulsen)
(DREAM working paper, 2021)
In this paper, we estimate the elasticities of substitution of imports and exports used in MAKRO for the industries manufacturing and energy. The data used is from the international trade database BACI and is available on annual frequency for the period 1995-2016. By using detailed trade data for many countries at the product level, the trade elasticities can be estimated while taking into account typical endogeneity problems and it allows potential problems with aggregation bias to be investigated and addressed. Our methodical starting point is Feenstra (1994), where the supply and demand curves are separately identified by utilizing the heteroskedasticity of shocks in the country dimension. The resulting import elasticity between domestic and foreign-produced manufacturing goods is estimated at 2.76. We then estimate the export elasticity as a weighted average of the import elasticity of Denmark’s 50 largest trading partners and obtain an estimate of 5.42 for the manufacturing sector. This estimated export elasticity is robust to an alternative assumption, that the elasticity is the same across the countries to which Denmark exports. For energy, we find elasticities of 3.50 and 5.03 for imports and exports, respectively. There are indications of aggregation bias, as the elasticity estimates are generally higher when more disaggregated data is used in the estimation.
Supply effects in Danish Exports?
(With Anders F. Kronborg)
(DREAM working paper, 2021)
In this paper, we examine to what extent supply effects are present in data for Danish exports. We apply detailed trade data for the time period 1995-2016. First, we analyze to what extent variations in expots are driven by the intensive and extensive margin. We find, that the intensive margin is the main determinant of short-run fluctuations in exports, and long-run fluctuations are determined by the extensive margin.
We subsequently estimate the supply effect of Danish exports based on a set of Gravity equations. We apply the actual variables, their structural levels, and instruments to remove business cycles and address endogeneity problems. Our preferred estimate of the supply effect for private Gross Value-Added indicates that the elasticity is not significantly different from unity. Finally, we find indications that the supply variables create a scale effect via the extensive margin. We conclude that supply effects are significant in the description of Danish exports, especially in the long run, and should be included in the specification of exports in MAKRO.