The Economic and Political Effects of Espionage

Project overview

Economic and political espionage are pervasive phenomena around the world. While countries like China and Russia are widely believed to engage in large-scale economic espionage, the practice seems to be much more widespread, with even countries formally considered allies collecting information about each other to give their economies a competitive edge in the global market. 

Due to its secretive nature, reliable data on espionage are difficult to come by. Therefore, this project exploits historical data from the Cold War period during which the East German Ministry for State Security (referred to as Stasi) engaged in comprehensive espionage in the Western world, especially West Germany. 

This project focuses on three work packages which have the following objectives: 

1. To deliver a quantitative assessment of the gains from economic espionage and to uncover the mechanisms through which firms assimilate espionage-based scientific-technical information
2. To shed light on the political effects that espionage might have on the affected countries
3. Understand the factors driving an individual’s decision to work for the Stasi

Summary, output, and dissemination

Research summary

Work Package 1

After successfully linking the espionage inflow information from SIRA with the firm-level information from the East German Statistical Office, this project estimated a comprehensive set of regressions that link changes in different firm-level outcomes to the inflow of espionage-based information these firms received over the previous period. 

The sample consists of 142 East German combines, which was observed over the period from 1985 to 1989, in roughly five-year intervals. A positive effect was found between the espionage inflows on firms’ output, exports, productivity, material investments, and the actual working time of the workforce: a 10 percentage point increase in the inflow rate increases output in the following five years by 1.0 percent, exports by 0.8 percent, productivity by 1.0 percent, material investments by 3.3 percent, and actual working time per worker by 0.04 percent. 

On the other hand, no evidence was found of any statistically significant effect of espionage inflows on the total capital stock, employment, the skill mix of workers, and the overall wage bill. The project systematically tested the robustness of the findings to alternative specifications, including one where the control for firm-level patenting and different functional forms of the variables of interest. An instrumental variables approach was also applied to deal with the potential endogeneity of espionage inflows (which led to qualitatively similar results). 

Out of the extensive set of results obtained, the positive impacts on productivity and export activities are probably the most noteworthy ones as they indicate that the inflow of espionage-based scientific-technical information allows firms to raise their productivity and to improve their competitiveness on the world market (as reflected by the positive impact on exports). The finding that espionage inflows raise investment, but not overall employment or the skill mix of workers, further suggests that the type of knowledge obtained is primarily complementary with other capital inputs but less so with different types of labor. 

To complement these findings, novel data was collected on East German firms’ privatization outcomes after German unification over the last few months. Preliminary results based on this data show that East German firms that received more espionage-based information during the Cold War period are more likely to be privatized, another clear indicator of economic espionage helping firms to keep up with the technological frontier

Work Package 2

The first step for this part of the project was to analyze the SIRA data and understand the full extent of the Stasi’s political espionage in the West. Figure 1 shows the annual inflow of corresponding information as recorded in the SIRA system:

Figure 1: Annual inflow of information related to political espionage between 1968 and 1989 

The content of each piece of information in the SIRA system is described by a set of keywords. Figure 2 shows the most frequently occurring keywords between 1951 and 1989. Leading the list, the word “Objekt” (Object) was used to label 53% of the incoming pieces of information over this time period. It is followed by “Name” (Name), “Außenpolitik” (Foreign Policy) and “Innenpolitik” (Domestic Policy), which were used to label 28%, 24% and 15% of all incoming pieces of information, respectively. Among the list of frequent keywords are further interesting ones such as “Partei” (Political Party), “Streitkräfte” (Armed Forces), “Ost-West Beziehungen” (East-West Relations) and “Wahl” (Election).

Figure 2: Most frequently occurring keywords in the realm of political espionage 

To study further the extent to which the Stasi was able to infiltrate West German political parties, the researchers computed the number of pieces of information that explicitly referred to one of West Germany’s parties. Figure 3 shows the corresponding time series over the sample period.

Figure 3: The time series of the frequency for incoming information with party-related keywords

Unfortunately, this project has not been able to link these data in a reasonable way to any relevant outcome variables, therefore the work so far has been primarily descriptive in nature. The problem is the lack of cross-sectional variation which makes it hard to isolate the impact of political espionage from other changes in the outcome variables of interest over time. For example, the researchers looked at West German citizen’s survey responses about their attitudes towards East Germany and correlated those responses with a measure of the East’s political espionage intensity (since one of the main objectives of the East’s political espionage was to improve the West German population’s perception of the GDR). However, from this simple correlation over time, it was hard to draw any meaningful conclusions. The researchers have not given up on this work package, focusing especially on the infiltration of different media outlets and how that might have affected the sentiment of newspaper articles towards East Germany. This research was impacted by the disruption caused by the Covid-19 pandemic, but the research continues.

Work Package 3

Based on the newly-created linked data, we have looked in detail at the way in which spies’ productivity, both in terms of quantity and quality, evolves over their career with the Stasi. If there is some learning involved, one would expect spies to become more productive in obtaining relevant information the longer they work for the Stasi. Figure 4 depicts how the average number of pieces of information per year develops over the spies’ “tenure” with the Stasi. As expected, there is a clear concave pattern reminiscent of standard age-earnings profiles, indicating that spies initially raise their productivity substantially, but that this process of learning is completed after around 10 years into their careers. Figure 5 shows that productivity increases not just in terms of the pure volume of information delivered, but also the average quality of this information (where 5 is the highest possible value and 1 the lowest).

Figure 4: Annual pieces of information vs. tenure as spy 

 

Figure 5: Annual pieces of information vs. tenure as spy 

This project extracted biographic information on individual spies from the existing literature, which was used to identify the spies in the West German social security records. The researchers are now in the process of preparing the data for the first set of estimations. 

Overall, despite the difficulties that this project faced during the pandemic, the researchers are optimistic that they will be able to generate intriguing new insights into how spies actually go about their trade, in the near future.

Prizes and recognition

“Europa Excelencia” Grant (EUR2021-122012) by the Spanish Ministry of Science and Innovation for a follow-up project on the economics of espionage, 2021

Publications

• Industrial Espionage and Productivity (2020), with Erik Meyersson. American Economic Review, 110 (4), 1055-1103
• Occupational Recognition and Immigrant Labor Market Outcomes (2021), with Herbert Brücker, Adrian Lerche and Agnese Romiti, Journal of Labor Economics, 39 (2), 497-525
• Learning Through Coworker Referrals (2021), with Rune Vejlin, Review of Economic Dynamics, 42, 37-71
• Skill Premiums and the Supply of Young Workers in Germany (2021), with Daniel Wissmann, Labour Economics, 72, 102034

Dissemination and related activities

• Presentation of the paper “Espionage-based Knowledge Flows and Firm Outcomes” (Albrecht Glitz, Adrian Lerche, Lukas Mergele) at the Annual Conference of European Association of Labor Economists in Padova, September 2022
• Presentation of the paper “Espionage-based Knowledge Flows and Firm Outcomes” (Albrecht Glitz, Adrian Lerche, Lukas Mergele) at the Annual Conference of the Verein für Socialpolitik in Basel, September 2022
• As soon as full drafts are available, the research will be submited to widely accessible working paper series such as the CEPR Discussion Paper Series, the IZA Discussion Paper Series, and the Barcelona School of Economics Discussion Paper series. We will also continue presenting our work at international conferences.