Economic Impact of immigration in Germany

 The sources introduce a unifying analytical framework that extends the canonical model of immigration's impact on labor markets,. This framework is designed to bridge the gap between "places" (regional outcomes) and "people" (worker-level outcomes), arguing that traditional estimates based on repeated cross-sectional data provide only composite effects that bundle distinct, unidentified mechanisms,.

The Core Framework: Places vs. People

The central argument of the framework is that changes in regional outcomes do not necessarily reflect the experiences of individual workers within those regions,. To address this, the sources provide a systematic decomposition of regional effects into their underlying components:

• Employment Effects: The regional employment effect is decomposed into three channels:
  • Displacement Effect: The impact on the employment prospects of natives already working in the region when immigrants arrive,.
  • Crowding-out Effect: Native workers who are discouraged from entering the labor market in the affected region,.
  • Relocation Effect: Native workers who respond to immigration by moving to employment in other regions.
• Wage Effects: The regional wage response is decomposed into:
  • Pure Wage Effect: The decline in wages caused by an outward shift in the labor supply curve along a downward-sloping demand curve,.
  • Compositional Changes: Changes in regional wages driven by shifts in the "quality" or "productive efficiency" of the native workforce (e.g., if low-productivity workers leave the region, the average wage might appear stable even if the price of labor has fallen),,.

Theoretical Foundations

The framework incorporates individual heterogeneity in two key ways:

1. Productive Efficiency: Workers have different efficiency levels, and their log wage is a combination of a time-constant "worker fixed effect" and the regional price of labor,.
2. Labor Supply Elasticities: Different groups of natives adjust their labor supply differently in response to immigration,.

The sources distinguish between a population-weighted labor supply elasticity (reflecting total headcount) and an efficiency-weighted aggregate labor supply elasticity. A "selectivity bias" arises in regional data if these two elasticities differ; for instance, if low-productivity workers are more likely to leave employment due to immigration, the average quality of the remaining workers improves, which can mask the negative "pure" wage effect,.

The Role of Longitudinal Data

A primary contribution of this framework is showing that these underlying components can only be identified using longitudinal data that track individuals over time,. While cross-sectional data only show regional averages, longitudinal data allow researchers to:

• Identify the pure wage effect by looking at individual wage growth for "stayers" (natives who remain employed in the same region),.
• Assess the impact on understudied groups, such as young labor market entrants or older workers who may be pushed into early retirement,,.
• Correct for selectivity bias without relying on complex structural models or instrumental variables for every component,.

The sources emphasize that the impact of immigration on employment must be understood by distinguishing between regional outcomes ("places") and individual worker outcomes ("people"). While traditional cross-sectional studies measure the "regional employment effect," the sources argue this is a composite parameter that bundles together three distinct mechanisms: displacement, crowding-out, and relocation.

Decomposing Employment Impacts

The analytical framework provided in the sources decomposes the total change in native regional employment into the following channels:

• Regional Employment Effect: This measures the overall change in native employment in a region hit by an immigration shock. In the empirical application involving Czech commuters to Germany, a one percentage point increase in the immigrant share reduced regional native employment by 0.873 percent.
• Displacement Effect: This refers to the impact on the employment prospects of natives already working in the region when the shock occurs. The sources find that this effect is relatively modest (increasing the probability of job loss by only 0.139 percent) and often disappears within five years.
• Crowding-out Effect: This occurs when native workers who were not previously employed in the region are discouraged from entering its labor market. This was found to be the primary driver of the regional employment decline, accounting for 88 percent of the total regional effect.
• Relocation Effect: This measures the extent to which employed workers move to other regions in response to immigration. In the study's application, this effect was found to be small and statistically insignificant.

Heterogeneity Across Worker Groups

The employment impacts of immigration are not distributed evenly across the native population. The sources highlight several groups that bear a disproportionate burden:

• Older Workers: Natives aged 50 and above experience a much stronger displacement effect. A one percentage point increase in the Czech employment share increased the probability of older workers being no longer employed by 1.145 percentage points.
• Non-Employed Workers: Those actively seeking work at the time of the shock are more affected than those already in jobs. The displacement effect for this group was 0.405 percentage points, significantly higher than the effect on incumbents.
• Routine vs. Abstract Workers: Native routine employment falls sharply in response to routine-biased immigration, while abstract employment remains stable. However, the sources find no evidence of individual "upgrading" (workers moving from routine to abstract jobs); instead, the decline in routine employment is again driven by reduced inflows (crowding-out).

Impact on Educational Choices

The sources also reveal a secondary employment-related response among young job market entrants. Instead of entering the labor market directly—where competition with low-skilled immigrants is highest—native school leavers are more likely to pursue vocational training. Specifically, a one percentage point increase in the immigrant share raised native apprenticeship take-up by approximately 1.3 percent.

The sources highlight a significant discrepancy between regional wage effects (the impact on a location) and pure wage effects (the impact on individual workers), arguing that traditional cross-sectional data often fail to capture the true economic consequences of immigration for native workers,.

The Distinction Between Regional and Pure Wage Effects

The analytical framework provided in the sources decomposes the impact on wages into two primary channels:

• The Pure Wage Effect: This represents the change in the price of labor caused by an outward shift in the labor supply curve along a downward-sloping demand curve,. In the studied German border region, a one percentage point increase in the immigrant employment share resulted in a negative pure wage effect of 0.188 percent after three years and 0.249 percent after five years,.
• The Regional Wage Effect: This measures the average wage change across a specific geographic area using cross-sectional snapshots. In the sources' application, this effect appeared close to zero (-0.008 percent in 1993), which would misleadingly suggest that immigration had no impact on native earnings,,.

The Role of Selection and Workforce Composition

The reason for the gap between regional and individual outcomes is selectivity bias or changes in the "quality" of the remaining native workforce,. The sources explain that immigration can induce changes in workforce composition that mask the decline in the price of labor,:

• Improved Average Quality: If low-productivity workers (who often have higher labor supply elasticities) are more likely to leave the labor market in response to immigration, the average "productive efficiency" of the remaining workers in that region rises,,.
• Offsetting Mechanisms: In the German application, this improvement in worker quality almost exactly offset the negative pure wage effect, causing regional averages to remain stable even as individual wages for "stayers" fell.
• Inadequacy of Observables: The sources note that simply controlling for observed characteristics like age or education is insufficient to isolate the pure wage effect, as quality changes also occur within these specific groups,.

Heterogeneity and Understudied Groups

Wage impacts vary significantly depending on a worker's initial status:

• Non-Employed Workers: Natives who were not employed at the time of the shock but were seeking work experienced much larger wage losses than those already in jobs,. Their pure wage effect was -0.706 percent in 1993, potentially due to skill depreciation during increased unemployment durations caused by competition with immigrants,.
• Occupational Differences: While Czech immigrants primarily entered routine occupations, the pure wage effects were found to be relatively similar across both routine and abstract occupations (-0.187 percent versus -0.121 percent).

Methodological Implications for Economic Elasticity

Using the wrong wage parameter leads to "implausible" calculations of key economic indicators,. When the sources used the pure wage effect identified from longitudinal data, they inferred a labor demand elasticity of -0.51, which aligns with established economic meta-studies,. In contrast, basing calculations on the regional wage effect resulted in an extreme and "extreme" demand elasticity of -12.5, demonstrating the necessity of tracking individuals to understand market dynamics.

To visualize this, imagine a professional basketball team that recruits several new players, causing the team's average height to stay exactly the same. At first glance, you might think recruitment didn't change the team's height profile. However, if you track the individual players, you might find that the veterans actually lost half an inch of height due to aging (the pure effect), but the team's average remained stable because the shortest players were the ones cut from the roster (the composition effect). Without tracking individual players over time, you miss the reality that every remaining player is actually shorter than they were before.

The sources emphasize that the impact of immigration is not uniform; instead, it triggers heterogeneous responses across different segments of the native population based on their employment status, age, and occupation,. Within the "Places versus People" framework, these varied responses explain why regional averages often fail to capture the specific hardships or adaptations of individual worker groups,.

Impact on Non-Employed vs. Employed Workers

The sources highlight that natives who are actively seeking work (non-employed) at the time of an immigration shock bear a significantly higher burden than those already in stable jobs,.

• Wage Losses: Three years after the immigration shock, previously non-employed workers experienced a pure wage effect of -0.706%, which is substantially larger than the -0.188% found for employed natives,.
• Displacement: The probability of failing to find employment was 0.405 percentage points for this group, nearly triple the displacement effect felt by incumbent workers,. This may be due to skill depreciation caused by longer unemployment durations as they compete with immigrants for available openings.

Vulnerability of Older Workers

Older workers (aged 50 and above) are identified as a particularly sensitive group,.

• Increased Displacement: A one percentage point increase in the immigrant share raised the probability of an older native worker leaving employment by 1.145 to 1.169 percentage points,.
• Compositional Influence: Because older workers often earn higher wages, their exit from the labor market significantly impacts the regional workforce composition, potentially dragging down average regional wages even if the individual "price" of labor for other groups remains stable,.

Routine vs. Abstract Occupations

Because the immigration shock studied in the sources was heavily routine-biased (95.1% of Czech commuters entered routine or manual jobs), the native response was highly lopsided,.

• Employment Shifts: Native routine employment fell sharply, while abstract employment remained stable.
• The "Upgrading" Myth: While regional data showed an increase in the share of natives in abstract jobs, the sources found no evidence of individual upgrading (adults moving from routine to abstract roles),. Instead, the shift was caused by crowding-out, where new native workers were deterred from entering routine roles, and older routine workers retired.

Educational Adaptation in Young Entrants

While adult workers rarely "upgraded" their skills, native school leavers showed a clear heterogeneous response by opting for more education to avoid low-skilled competition,.

• Apprenticeship Take-up: The sources found a sharp increase in native apprenticeship enrollment (1.3% to 1.4%) following the immigration shock,. This suggests that the primary "upgrading" mechanism in response to immigration occurs at the point of labor market entry rather than through mid-career transitions,.

Theoretical Drivers: Efficiency and Elasticity

The framework explains these heterogeneous outcomes through two variables: productive efficiency and labor supply elasticity,.

• Selectivity Bias: Lower-productivity workers typically have higher labor supply elasticities, meaning they are more likely to leave the workforce or a specific region when wages drop due to immigration,.
• Masking Effects: Because these lower-paid workers leave the "place" (the regional data), the remaining workforce appears to be of "higher quality" or "higher efficiency," which masks the negative pure wage effects experienced by those individuals who remain,.

To understand these heterogeneous responses, imagine a department store that hires many new seasonal staff for the gift-wrapping department (routine tasks). The veteran managers (abstract workers) aren't replaced, but the older employees near retirement might decide to leave earlier because the floor is more crowded and stressful. Meanwhile, the local teenagers who usually apply for gift-wrapping jobs see the new hires and decide to go to community college instead of applying. A census of the store might show that the average employee is now more highly "educated" (because only the managers and students stayed), but this masks the fact that the actual gift-wrappers are earning less and the local seniors have lost their jobs.

The sources argue that traditional methodologies used to study immigration—primarily those relying on repeated cross-sectional data—provide an incomplete picture because they only identify composite effects that bundle distinct economic mechanisms together. To address this, the authors propose a unifying empirical framework that leverages longitudinal data to track individual workers over time, allowing for a precise decomposition of regional impacts into specific worker-level outcomes.

The Identification Strategy: Natural Experiment and IV

The methodology is built around a specific policy experiment in Germany that allowed Czech workers to commute to German border regions for work without granting them residency. This provides a clean identification strategy for several reasons:

• Pure Labor Supply Shock: Because commuters lived in the Czech Republic, they did not significantly increase local demand for goods, isolating the impact to the labor supply side.
• Instrumental Variable (IV) Approach: To account for the fact that immigrants might choose to work in booming municipalities, the researchers use the municipality’s distance to the border (and its square) as an instrument. This assumes that distance predicts immigrant inflow but does not otherwise correlate with native employment or wage trends.
• Event Study Design: The authors estimate effects both "forward" (after the shock) and "backward" (before the shock) to ensure that affected and unaffected regions were following common trends prior to the policy change.

Decomposition of Regional Effects

A major methodological contribution of the sources is the systematic decomposition of regional averages into identifiable components:

• Employment Decomposition: Regional employment changes are broken down into displacement (loss of existing jobs), crowding-out (reduced entry of new workers), and relocation (workers moving to other regions).
• Wage Decomposition: Regional wage changes are decomposed into the "pure" wage effect (the shift in the price of labor) and compositional changes (shifts in the average quality/efficiency of the workforce).

The "Within-Estimator" and Addressing Selection Bias

To isolate the true economic impact on individuals, the sources utilize a within-estimator at the individual level.

• Individual Wage Growth: By restricting the sample to "stayers"—natives who remain employed in the same municipality—the methodology identifies the pure wage effect under the assumption that selection out of the region is driven by time-constant individual characteristics.
• Correcting for Unobservables: The framework eliminates "worker fixed effects" (time-constant productivity) through differencing. To handle potential bias from time-varying unobservables, the authors adapt a bounding approach, which helps quantify the maximum possible influence of unobserved factors on the results.
• Pseudo-Panel Limitations: The sources demonstrate that even a pseudo-panel approach (adjusting for observed traits like age and education) is insufficient, as immigration triggers compositional changes within those specific groups that only longitudinal tracking can detect.

Theoretical Grounding: The Extended Canonical Model

The methodology extends the canonical supply-and-demand model by incorporating individual heterogeneity in two dimensions: productive efficiency and labor supply elasticity. This allows the researchers to distinguish between a population-weighted elasticity (headcount) and an efficiency-weighted elasticity (productivity), which is crucial for identifying the "selectivity bias" that often masks the negative impacts of immigration in regional data.

To grasp this methodology, imagine trying to understand the health of a school of fish after a change in the water current. A traditional cross-sectional study is like taking a photo of the school before and after; you might see the school is the same size, but you won't know if the original fish were replaced by different ones. This longitudinal methodology is like tagging every individual fish. By tracking each one, you can see if the original fish are growing slower (pure wage effect) or if the smaller, weaker fish left the school and were replaced by larger ones, making the school look "healthy" on average while the individuals are actually struggling.

In the analytical framework provided by the sources, crowding-out refers to the reduction in native workers entering a local labor market in response to an immigration shock. It represents "missing inflows" from either non-employment or from workers living in other regions who would have otherwise sought jobs in that area.

The sources identify crowding-out as the primary driver of regional native employment declines, playing the following specific roles:

1. The Primary Mechanism for Regional Decline

While public debate often focuses on "displacement" (incumbent workers losing their jobs), the sources find that crowding-out accounts for the vast majority of the impact on regional employment. In the studied German border region:

• A one percentage point increase in the immigrant share reduced regional native employment by 0.873 percent.
• Crowding-out was responsible for 88 percent of this total regional effect.
• In contrast, the actual displacement of workers who were already employed in the region was modest (0.139 percent) and disappeared entirely after five years.

2. Decomposition of "Missing Inflows"

The sources further decompose the crowding-out effect to show where these "missing" native workers originated. Three years after the immigration shock, a one percentage point increase in the immigrant share reduced the inflow rate by 0.768 percentage points, which consisted of:

• Reduced inflows from non-employment: This accounted for approximately two-thirds of the crowding-out effect.
• Reduced inflows from other regions: This accounted for the remaining one-third, where native workers employed in other municipalities were deterred from moving to the affected region.

3. Influence on Occupational and Educational Choices

Crowding-out is particularly prevalent in routine and manual occupations, where the studied immigrants were most concentrated.

• Deterrence of Routine Entrants: The decline in native routine employment was driven almost entirely by reduced inflows (crowding-out) rather than adult workers "upgrading" to abstract jobs or being displaced.
• Educational Substitution: Crowding-out often manifests as young job market entrants choosing alternative paths. Instead of entering low-skilled employment and competing with immigrants, native school leavers were found to be more likely to enroll in apprenticeship training schemes, with a one percentage point increase in immigrant share raising apprenticeship take-up by 1.3 to 1.4 percent.

4. Methodological Significance

From a methodological standpoint, the role of crowding-out explains why repeated cross-sectional data can be misleading. Such data show a drop in the number of native workers in a "place," but without tracking individuals (longitudinal data), researchers cannot distinguish between people losing their jobs (displacement) and people simply never arriving (crowding-out).

The sources note that crowding-out has ambiguous welfare implications. Unlike displacement, which affects identifiable individuals who suffer direct job loss, the individuals who are crowded out are not identifiable, and it is possible they simply found equally good employment in regions less affected by immigration.

---

To visualize this, imagine a popular local bakery that suddenly hires several new workers from a neighboring town. A "regional" study sees fewer local residents working there and assumes the locals were fired. However, tracking the individuals reveals that the veteran bakers all kept their jobs. The decline in local staff happened because the neighborhood teenagers, who usually apply for their first jobs at that bakery, saw there were no openings and decided to stay in school or find work at a shop three towns over. The bakery's "employment decline" for locals was caused by a closed door (crowding-out) rather than a pink slip (displacement).