Abstract
This paper estimates the long-term impacts of schooling disruptions on private returns to schooling in Kuwait. It applies an instrumental variables approach to estimate the private returns to schooling, using unique civil service payroll data, with Kuwaiti students’ exposure to the Gulf War (1990–91) as the instrument. The Gulf War is a suitable instrument because it profoundly affected Kuwaiti students' schooling at the time and is unlikely to be correlated with many potentially problematic omitted variables, such as students’ ability. The analysis finds that (i) people who were of schooling age during the Gulf War tend to have lower educational attainment than people who were of schooling age after the Gulf War; (ii) men who were of schooling age at the time of the Gulf War earn on average 5.6% less for each year of schooling lost, and women earn correspondingly 6.8% less for each year of schooling lost; (iii) female students who were in the age groups corresponding to lower school grades during the Gulf War tend to suffer a greater percentage wage loss for each year of lost schooling.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data Availability
The payroll data from the Kuwait Civil Services Commission used in this study are confidential and cannot be made publicly available.
Notes
We use the term “Gulf War” broadly, including the initial Iraqi invasion and occupation of Kuwait and the resulting military campaign of a coalition of countries against Iraq.
Our analysis also connects to a broader literature exploring the effect of conflict on labor market outcomes. Kondylis (2010), for example, finds adverse effects of conflict-related displacement on labor market outcomes in post-war Bosnia and Herzegovina. Similarly, Morales (2018) finds adverse labor market effects of incoming conflict-induced migration in Colombia, where the effects tend to dissipate over time.
Dang et al. (2021) discuss an interesting case of the First Indochina War (1946–1954) associated with an increase in girls’ education outcomes in areas controlled by the Democratic Republic of Vietnam. However, the positive association results from the education policies in the areas, not the war itself. Similarly, Pivovarova and Swee (2015) find no impact of war intensity in Nepal on schooling attainment, suggesting the impact of egalitarian Maoist ideology might be counteracting the adverse effects of the war.
Justino et al. (2013) present a nuanced analysis of the consequences of the 1999 wave of violence in Timor Leste on primary school completion. While affected girls had a higher chance of completion in the medium run, boys’ probability of completing primary school decreased.
The invasion began during the summer break of 1990/1991. Many families had already left the country for their summer vacation and decided to stay there as the invasion was unraveling. By one account “more than half the population, expats and citizens alike, were out of the country” (Manatt and Manatt 1996: 368).
Our sources do not state whether private schools followed suit, however, private schools at the time served a smaller, although substantive, percentage of students in Kuwait. In 1991/92, 245,800 students were studying at pre-tertiary public schools, including kindergartens (Central Statistical Bureau 2000: 311). In comparison, there were 74,784 students attending private schools. The data include Kuwaitis and non-Kuwaitis as we were not able to find comparable data disaggregated by nationality. Al-Qudsi (1989: 266) notes that 99% of Kuwaiti nationals were attending public pre-tertiary level education but does not provide a specific source for the information.
In the report from August 1991, the Kuwaiti government decided to dismiss 110,000 civil servants, mostly Palestinians, which led to recruitment drives, including those for teachers. The dismissals happened partly because of the perceived collaboration of the foreign workers with the Iraqi regime. The dismissals of the Palestinians were also “a retaliation for the Palestine Liberation Organisation's support for Iraq” during the war (Isa 1991).
Note that while, generally, there should be a correspondence between a student’s age and the grade in which this student enrolls, it does not have to be the case. For details, see Appendix 1 for the age ranges for each grade level.
Different rules might apply to private schools. At the private American School of Kuwait, first-year elementary students, for example, have to be six years old by December 31 (Admissions Requirements n.d.).
1 Kuwaiti dinar was equivalent to US$3.33 on April 6, 2021.
It should also be noted that the compensation schedule does not reflect one’s age group, which is important for exclusion restriction assumption in our analysis later. See Alqattan et al. (2012) for a translated version of the compensation schedule.
Wage data in household and labor force surveys are often underreported and suffer from measurement errors.
One can make similar objections for those listed as Kuwaiti nationals. Some may have been living outside Kuwait before the Gulf War, and their lives might have been disrupted only minimally by the war. Although we do not have the exact number of such nationals, and as there are likely very few, the absence of the information would be unlikely to substantially bias the results.
One might object to not utilizing the non-Kuwaiti segment of the data as it might serve as a control group when analyzing the Gulf War's impact. While it is true that the schooling of many non-Kuwaitis was likely unaffected by the war, they also represent a heterogeneous population that might substantially differ from Kuwaitis. An interpretation of a comparison of the two might then be difficult because the differences might not amount to the effects of the war but rather to the underlying differences between Kuwaitis and non-Kuwaitis. Moreover, the population of non-Kuwaitis working for the Kuwaiti civil service is likely endogenous as their hiring often tends to be conditional on the absence of Kuwaiti candidates. If we used the employed non-Kuwaitis as a control group to analyze the Gulf War's impact on Kuwaitis, we would have a situation with an endogenously selected control group and biased results.
There were 408,205 employed Kuwaitis in Kuwait in December 2019, of which 344,937 were working in the public sector according to PACI (2019). Keeping in mind that our data was collected about 6 months earlier than PACI’s, the data covers approximately 76% of all Kuwaiti public sector workers and 64% of all Kuwaiti workers. Note that the data we use does not represent all public sector employees as the Civil Services Commission does not hire all public sector workers in Kuwait.
Public and private sectors are connected through a wage arbitrage; however, public sector compensations have the “upper hand” here. The public sector’s appealing compensations and a tendency to employ the excess supply of Kuwaiti nationals, both discussed earlier, have the tendency to function as a price floor for the private sector. The floor’s mechanism then tends to distance a Kuwaiti national’s public sector salary from the corresponding marginal product of labor and wage that this person would make in the private sector.
Four years for primary school, four years for middle school, and four years for high school.
Note that the achieved education level here differs from the education-level age groups in Table 2. The former reflects one’s education achievement, the latter reflects one’s age.
In other words, we do not subtract the schooling time missed during the Gulf War from one’s overall years of schooling. For example, if a person finished high school and had one year off during the war, we still count it as 12 years of education—the same as any high school graduate—and not 12 years minus the time lost. We would subtract the missed time in an ideal case, but we do not have enough individual-level information. For example, if someone is a high school graduate and studied during the war, it can mean two things. First, he could have missed schooling because of the war. Second, he could have been studying abroad during the war, enrolling in a private school after the war, missing no schooling, and having 12 years of education. Unfortunately, we do not know which of the two applies in the case of a particular person. In other words, the lost schooling that we estimate in the paper is happening only after the war.
Someone born on March 16, 1992 likely enrolled in first grade in September 1997, at the age of 5 and a half. If this person eventually decided to get a PhD, the person would have graduated in 2019, after 22 years of schooling that we assume it to take in Appendix 2. It seems possible but not likely that under such circumstances, the person would be already hired by the civil service and thereby reflected in our data set from June/July 2019. All the people younger than March 16, 1992 are even less likely to have graduated, and therefore less likely to be included in our data set of active civil service employees, skewing the average years of schooling for the younger group downward.
Using data from Table 3 for men: 13.38–13.99 = − 0.61. Using the same source for women: 14.18–14.77 = − 0.59.
See Appendix 6 for a more detailed discussion of the treatment and control groups.
While Kuwait does not impose income taxes, employees are subject to a social security deduction. The net wages are gross wages minus the deduction.
Years of service are measured as years of work for Kuwait’s civil service.
This means that the number of children is nonzero only when the person receives children benefits. Since the benefits are usually collected by men on behalf of the family, most women have number of children listed as zero.
We do not have a post-graduate group for women because of the more restrictive age constraint. The treatment group for women includes only females born between January 01, 1971 and March 15, 1985. The male treatment group also includes men born between January 1, 1966 and March 15, 1985, reflecting the gender differences in the retirement rules.
For example, one can imagine a setting with two students, A and B. Let’s say that without a war, student A would achieve 16 years of schooling and student B 18 years. Assume further that with the war, student B ends up with 14 years of schooling, and A still ends up with the same 16 years. In this case, a LATE approach allows estimating only private returns to the schooling of student B—a student whose outcome changes because of the war.
The following discussion of the assumptions adopts Angrist and Pischke (2008: ch.4).
Kesternich et al. (2014) find negative long-term physical and psychological health effects in people exposed to World War II. One can still argue, however, that these effects did not have substantial consequences on long-run labor market outcomes.
Our dataset classifies employees into 19 functional groups, such as engineering jobs and administrative support jobs. We computed relative frequency distributions for the treatment group and the control group by gender, finding the Pearson correlation coefficient between the two distributions to be 0.94 for women and 0.91 for men. The data also classifies employees by the employing organization, like the Ministry of Education or Ministry of Finance. Analogous Pearson correlation coefficients are 0.996 for women and 0.91 for men.
Four for females.
Again, all post-estimation diagnostic tests reject the exogeneity of the years of education variable and reject the hypothesis of our instruments' weakness, thereby giving credibility to our methodological approach.
For example, it is possible that the war somehow dissuaded particularly productive fifth grader males from studying as many years as they would have otherwise.
Lost years of schooling are based on the estimates from Table 14 in Appendix 6 (-0.393 years for females, -0.496 for males), private returns to schooling are retrieved from Table 8 (8.6% for females, 8.8% for males). We compute the loss by multiplying the lost years of schooling and the private returns to schooling.
References
Admissions Requirements & Policies (n.d.) https://www.ask.edu.kw/admissions-requirements-and-policies. Accessed 1 Feb 2021
Ajwad MI, Koettl J, Radwan I, Farole T, Sanchez-Reaza J, Chartouni C, Alaref JJS, Rivera N, Sundararaman V, Afif Z, Dexter GA (2022) Towards a National Jobs Strategy in Kuwait. World Bank, Washington
Ajwad MI, Bilo S, Nguyen H, AlAnsari E, AlHumaidan L, AlRashidi F (2023) Female-worker representation effect: gender pay variation in the Kuwaiti Civil Service. Policy Research Working Paper; 10509, World Bank, Washington
Akbulut-Yuksel M (2014) Children of war the long-run effects of large-scale physical destruction and warfare on children. J Hum Resour 49(3):634–662
Akbulut-Yuksel M (2017) War during childhood: The long run effects of warfare on health. J Health Econ 53:117–130
AlAnsari EM (2018) Three essays about the Kuwaiti labor market: policies for labor market integration and pay differential based on gender and citizenship. Doctoral dissertation, American University, Washington DC
AlHumaidan L (2019) Domestic helpers and native female labor supply in Kuwait. Doctoral dissertation, George Washington University, Washington DC
Alqattan H, Stergioulas L, Al-Zayer J (2012) The returns to education based on level of education: case of Kuwait. European business research conference proceedings. https://ssrn.com/abstract=2130636. Accessed 22 Aug 2020
Al-Qudsi SS (1989) Returns to education, sectoral pay differentials and determinants in Kuwait. Econ Educ Rev 8(3):263–276
Al-Sahl R (1993) A study on the psychological and social effects of the Iraqi aggression on Kuwaiti children. Educ J 7(26)
Alsarraf Q, Alnaqeib K, Abu-Allam R, Alrefaei S (1993) دراسة مسحية حول الأوضاع التربوية و التعليمية في فترة ما بعد التحرير في دولة الكويت” [A survey study on the educational conditions in the post-liberation period in the State of Kuwait]. Kuwait Society for the Advancement of Arab Children
Angrist JD, Krueger AB (1991) Does compulsory school attendance affect schooling and earnings? Q J Econ 106(4):979–1014
Angrist JD, Pischke J-S (2008) Mostly harmless econometrics: An empiricist’s companion. Princeton University Press
Azevedo JP, Hasan A, Goldemberg D, Iqbal SA, Geven K (2020) Simulating the potential impacts of COVID-19 school closures on schooling and learning outcomes: a set of global estimates. World Bank Group
Bertoni E, Di Maio M, Molini V, Nisticò R (2019) Education is forbidden: the effect of the Boko Haram conflict on education in North-East Nigeria. J Dev Econ 141:102249
Brück T, Di Maio M, Miaari SH (2019) Learning the hard way: The effect of violent conflict on student academic achievement. J Eur Econ Assoc 17(5):1502–1537
Card D (2001) Estimating the return to schooling: Progress on some persistent econometric problems. Econometrica 69(5):1127–1160
Card D (1999) The causal effect of education on earnings. In: Ashenfelter O, Card D (eds) Handbook of labor economics, vol 3. Elsevier, pp 1801–1863
Central Statistical Bureau (2000) Annual Statistical Abstract. https://www.csb.gov.kw/Pages/Statistics_en?ID=18&ParentCatID=%202. Accessed 9 Feb 2021
Central Statistical Bureau (n.d.) The numerical and relative distribution of employment by gender and nationality. Web: https://lmis.csb.gov.kw/InteractiveReportPerSectorPerAge_En. Accessed 17 March 2022
Chamarbagwala R, Morán HE (2011) The human capital consequences of civil war: Evidence from Guatemala. J Dev Econ 94(1):41–61
Dang H-A, Hoang TX, Nguyen H (2021) The long-run and gender-equalizing impacts of school access: evidence from the first Indochina war. Econ Dev Cult Chang 70(1):453–484
Di Maio M, Nisticò R (2019) The effect of parental job loss on child school dropout: evidence from the occupied Palestinian territories. J Dev Econ 141:102375
Fergusson L, Ibáñez AM, Riano JF (2020) Conflict, educational attainment, and structural transformation: La violencia in Colombia. Econ Dev Cult Chang 69(1):335–371
Galdo J (2013) The long-run labor-market consequences of civil war: Evidence from the Shining Path in Peru. Econ Dev Cult Change 61(4):789–823
Garrison SB (2015) Kuwait female labor force participation: agency and development sustainability in Kuwait. Doctoral dissertation, Tulane University
Hadi F, Lai BS, Llabre MM (2014) Life outcomes influenced by war-related experiences during the Gulf crisis. Anxiety Stress Coping 27(2):156–175
Hanushek E, Woessmann L (2020) The economic impacts of learning losses. OECD Education Working Papers No. 225, OECD Publishing, Paris
Hosni DJ, Al-Qudsi SS (1988) Sex discrimination in the labour market of Kuwait. Int J Manpow 9(3):10–22
Ichino A, Winter-Ebmer R (2004) The long-run educational cost of World War II. J Law Econ 22(1):57–87
Imbens GW, Angrist JD (1994) Identification and estimation of local average treatment effects. Econometrica 62(2):467–475
Isa M (1991) Kuwait launches recruitment drive in Gulf Arab states. Reuters News, Middle East Economic Digest
Islam A, Ouch C, Smyth R, Wang LC (2016) The long-term effects of civil conflicts on education, earnings, and fertility: Evidence from Cambodia. J Comp Econ 44(3):800–820
Justino P, Leone M, Salardi P (2013) Short- and long-term impact of violence on education: the case of Timor Leste. World Bank Econ Rev 28(2):320–353
Justino P (2011) Violent conflict and human capital accumulation. MICROCON Research Working Paper 54, Institute of Development Studies at the University of Sussex, Brighton, UK
Kesternich I, Siflinger B, Smith JP, Winter JK (2014) The effects of World War II on economic and health outcomes across Europe. Rev Econ Stat 96(1):103–118
Kondylis F (2010) Conflict displacement and labor market outcomes in post-war Bosnia and Herzegovina. J Dev Econ 93(2):235–248
Kuwait Society for the Advancement of Arab Children (1994) Fall and resurrection: Kuwait and education- from independence to occupation and beyond. (The original title in Arabic: السقطة و القيامة: الكويت و التربية.. من الاستقلال إلى الاحتلال و ما بعدهما). (1994)
Kuwaiti Studies and Research Center (2002) The impact of the Iraqi invasion on the children of Kuwait "painful experiences". (The original title in Arabic: ” أثر الغزو العراقي على أطفال الكويت "تجارب مؤلمة”)
La Ganga ML (1991) Economy without a country Kuwait is conducting business without oil, its lifeblood. But with an estimated $100-billion portfolio, it is nowhere near broke. Los Angeles Times, February 22, 1991
Leon G (2012) Civil conflict and human capital accumulation the long-term effects of political violence in Perú. J Hum Resour 47(4):991–1022
Llabre MM, Hadi F (2009) War-related exposure and psychological distress as predictors of health and sleep: a longitudinal study of Kuwaiti children. Psychosom Med 71(7):776–83
Manatt RP, Manatt J (1996) After the gulf war—the schools of Kuwait. Int J Educ Reform 5(3):367–380
Middle East Economic Digest (1991) Schools open with help from US Army Corps of Engineers
Mincer JA (1974) Schooling, experience, and earnings. National Bureau of Economic Research, New York
Ministry of Education (2008) The national report: development of education in the State of Kuwait, 2004–2008. http://www.ibe.unesco.org/fileadmin/user_upload/archive/National_Reports/ICE_2008/kuwait_NR08.pdf. Accessed 18 Jan 2021
Morales JS (2018) The impact of internal displacement on destination communities: Evidence from the Colombian conflict. J Dev Econ 131:132–150
PACI (Public Authority of Civil Information) (2019) Approaching workers by sector, nationality (Kuwaiti / non-Kuwaiti) and gender. https://www.paci.gov.kw/stat/SubCategory.aspx?ID=1. Accessed 30 Jan 2021
Patrinos HA, Psacharopoulos G (2020) Returns to education in developing countries. In: Bradley S, Green C (eds) The economics of education. Academic Press, pp 53–64
PIFSS (The Public Institution for Social Security) (2017) PIFSS Guide. Public Relations & Information Department, The Public Institution for Social Security, Kuwait
PIFSS (The Public Institution for Social Security) (2019) Cases of retirement pension entitlement after amending the social insurance law by law no. (10) of 2019, as of 1/4/2019. https://www.pifss.gov.kw/Ar_Retire2019.cms. Accessed 30 Jan 2021
Pivovarova M, Swee EL (2015) Quantifying the microeconomic effects of war using panel data: evidence from Nepal. World Dev 66:308–321
Poirier T (2012) The effects of armed conflict on schooling in Sub-Saharan Africa. Int J Educ Dev 32(2):341–351
Psacharopoulos G, Patrinos HA (2018) Returns to investment in education: a decennial review of the global literature. Educ Econ 26(5):445–458
Registration in Public Schools (n.d.) (The original title in Arabic: التسجيل في المدارس الحكومية). https://www.e.gov.kw/sites/kgoArabic/Pages/Services/MOE/EnrollmentGovernmentSchools.aspx. Accessed 1 Feb 2021
Safwat I (1993) System of education in Kuwait. PIER World Education Series, Working Paper. https://files.eric.ed.gov/fulltext/ED399866.pdf . Accessed 11 Dec 2020
Sassoon J, Walter A (2017) The Iraqi occupation of Kuwait: new historical perspectives. Middle East J 71(4):607–628
Shemyakina O (2011) The effect of armed conflict on accumulation of schooling: results from Tajikistan. J Dev Econ 95(2):186–200
The Beginning of the Construction Process after Liberation: Tutoring, a Great Human Educational Achievement Achieved by the Kuwaiti Teachers Association in the Worst Conditions (1991) (in Arabic). Teach Mag 37–38
The Kuwaiti Educational Center in Cairo (1991) Draft of the proposed study plan for the post-return period. Ministry of Education
The Kuwaiti Teachers Association (1991) Academic Calendar for the Academic Year 1991/1992
UNESCO-IBE (2007) World data on education: sixth edition 2006–07. http://www.ibe.unesco.org/en/document/world-data-education-sixth-edition-2006-07. Accessed 18 Jan 2021
UNESCO-IBE (2012) World data on education: seventh edition 2010–11. http://www.ibe.unesco.org/en/document/world-data-education-seventh-edition-2010-11. Accessed 18 Jan 2021
Verwimp P, Van Bavel J (2013) Schooling, violent conflict, and gender in Burundi. World Bank Econ Rev 28(2):384–411
Weldeegzie SG (2017) Growing-up unfortunate: war and human Capital in Ethiopia. World Dev 96:474–489
Acknowledgements
This paper, its analysis, and conclusions are solely those of the authors and do not have to represent their respective organizations' views. The authors are grateful to Dr. Khaled Mahdi, Issam A. Abousleiman, Ghassan N. Alkhoja, and Anush Bezhanyan for their support. In addition, the authors are grateful to João Pedro Azevedo, Reyadh Faras, Maria Marta Ferreyra, Roberta V. Gatti, Laura Gregory, Johannes Koettl, Aart Kraay, Ha Nguyen, Harry Anthony Patrinos, Ismail Radwan, and participants of the Middle East and North Africa Chief Economist Seminar Series at the World Bank for helpful comments and valuable suggestions. Any remaining errors are the responsibility of the authors.
Funding
No funding was received to assist with the preparation of this manuscript.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of Interests
The authors have no relevant financial or non-financial interests to disclose.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix 1
Age Ranges that are Allowed for Registration at the Corresponding School Grade Level
The Ministry of Education allows students to be enrolled in a given grade at various ages. According to decision number (31,639) taken on the 17/5/2004, the following Table 9 shows the age ranges that are allowed for registration at the corresponding grade level (Registration in Public Schools n.d.).
Appendix 2
How We Mapped Years of Education to the Achieved Education Level Variable in the Data Set
Table 10 shows the underlying assumptions for our years of schooling variable.
Appendix 3
Employment Process at Kuwait’s Civil Services Commission
Figure 6 summarizes employment procedures at Kuwait's Civil Services Commission.
Kuwait’s Civil Services Commission employment procedures
Appendix 4
Minimum Retirement Age for Kuwaiti Civil Servants (Conditional on Fulfilling the Minimum Years of Service for Retirement)
Table 11 details the timetable of minimum retirement ages for women.
Table 12 details the timetable of minimum retirement ages for men.
Appendix 5
Employment of Kuwait Nationals by Sector, Gender, and Age
Table 13 shows that private versus public employment choice is similar for cohorts of proximate ages.
Appendix 6
Further Discussion on the Choice of our Treatment and Control Groups
To complement Table 3 in the main text, we present years of schooling in a more disaggregated way also in graphs. Figure 7 shows the average years of schooling for corresponding education-level and working-level age groups.
Kuwaiti civil servants’ average years of schooling in 2019, grouped by their estimated education- and working-level age groups in 1990
We estimate an OLS regression of the data represented in Fig. 7 with years of schooling as the dependent variable and independent categorical variables representing the various education-level age groups. The results in Table 14 confirm that the average years of schooling differ significantly between students who enrolled after the Gulf War and those of schooling age during the war.
The estimates in Table 14 and the information in Fig. 7 point to two further insights. First, unlike men, women's working age group is qualitatively similar to the education-level age groups. This likely results from the policy and cultural changes, that have made female education more accessible over time. Second, the average years of schooling of the working age male group are substantially above all the other groups. Figure 8 shows that these results are driven by older males, suggesting the average years of schooling might be biased by the retirement of older and less educated male workers. This is plausible, given the Kuwaiti civil service retirement policies we discussed above. These two insights support our decision to exclude from the analysis those people who likely finished their schooling before the Gulf War. The difference in years of schooling between this older group and younger groups is likely driven by causes other than exposure to the Gulf War. Since our analysis builds on the primary causal importance of the Gulf War in causing such differences, the inclusion of the older groups is not suitable for the analysis.
Kuwaiti civil servants’ average years of schooling. Grouped by year of birth (1955–2000)
Figure 8 also supports our decision to exclude people who are too young from the analysis. Male and female birthyear cohorts born after mid-1990s have declining average years of schooling, suggesting some members of these younger cohorts are yet to get more schooling. In other words, some of the prospective employees in the younger cohorts are still at school and will become Kuwaiti civil servants only in the future.
Appendix 7
2SLS Estimates of Private Returns to Schooling by Gulf War Grade and Post-War Control Group
Tables 15 and 16 present data used in Figs. 4 and 5.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Bilo, S., Ajwad, M.I., AlAnsari, E. et al. Estimating Long-Term Impacts of Wartime Schooling Disruptions on Private Returns to Schooling in Kuwait. J Labor Res 45, 111–152 (2024). https://doi.org/10.1007/s12122-023-09351-8
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12122-023-09351-8