Abstract
Income inequality in Hong Kong persists as a pressing concern, with the Gini coefficient consistently surpassing the global average. As income inequality has far-reaching implications, it is crucial to address this problem through different welfare programs. Education subsidy schemes are pivotal in bridging the gap, as they enhance educational opportunities for lower-income families. This study undertakes a cost-benefit analysis of various education subsidy schemes, assessing the increase in future income per government expenditure. By identifying the most cost-effective approach, this research provides recommendations for the government to optimize their education subsidy schemes, thereby mitigating income inequality in Hong Kong.
Table of Contents
- Introduction
- Methodology
- Result
- Discussions and Limitations
- Conclusion
Income inequality is a critical issue worldwide. This problem is not unfound in Hong Kong, with a notable surge in income and wealth disparities over the past four decades (Vu et al., 2022). The Gini coefficient of Hong Kong is forecast to amount to 0.47 in 2024 (Statista, 2024), passing the international inequality threshold alert line of 0.4. Income inequality has far-reaching implications, as it can result in a lack of economic opportunity for people coming from a lower-income family, perpetuating a cycle of financial hardship across generations.
To combat income inequality, the Hong Kong government has proposed welfare policies such as healthcare assistance (e.g. The Elderly Health Care Voucher Scheme), income subsidies (e.g. The Working Family Allowance Scheme) and public housing initiatives (e.g. Home Ownership Scheme). These policies also include schemes providing education subsidies. The effectiveness of educational subsidy schemes is most worth examining for three reasons. First, investing in education and allowing students to receive quality education is a major way of increasing future income and reducing income inequality. Chetty & Friedman (2011) estimate that the intergenerational correlation of income would fall by roughly a third if all children attended schools of the same quality, suggesting that differences in school quality perpetuate income inequality. By narrowing the gap in education quality between the rich and the poor, educational subsidy schemes can reduce their future income disparities. Also, educational subsidy schemes will likely increase future government revenue as future income increases. Hendren & Sprung-Keyser (2020) found that education policies amongst children have the largest marginal value of public funds (MVPF) when compared to other social welfare programs, meaning that it has the greatest impact on social welfare per dollar of government spending on the policy. This is because the increase in future government tax revenue offsets the costs of social welfare, resulting in an infinite MVPF
value for these subsidy schemes. The MVPF is calculated by dividing marginal benefits by net government costs. When net costs approach zero, the MVPF value approaches infinity. In other words, the expenditure is valued by beneficiaries and has no net long-run cost to the government. Moreover, educational subsidies allow underprivileged people to pursue their advanced education, possibly discovering talents to contribute to technology advancement that would otherwise be wasted on doing low-skilled jobs. Therefore, examining the effectiveness of different education welfare schemes is vital in reducing the income inequality problem and contributing to future economic growth in Hong Kong.
1.1 Current Literature
While few researches regarding the effectiveness of educational subsidy schemes in Hong Kong have been conducted, literature suggests a positive correlation between education inequality and income inequality (Abdullah, et al., 2013; Ahmed, 2023), emphasizing the need to reduce educational disparities to combat income inequality. However, as automation becomes prevalent, some research found that the effectiveness of education subsidies depends on the economy’s labour composition and employment status. In a fully employed economy, education subsidies tend to increase the wage gap between skilled and unskilled workers, especially when there aren't enough unskilled workers, as is the case in Hong Kong. Skilled workers often move to sectors that produce automation tools, boosting productivity there. This increases firms’ incentive to invest more capital in sectors that produce automation tools, reducing capital for unskilled labour and lowering their productivity. Hence, unskilled wages drop and the wage gap becomes wider. Conversely, in an economy with unemployment, education subsidies can help reduce the wage gap. As more workers gain skills, there may be a shortage of unskilled labour, which can lead to higher wages for unskilled workers (Zhang, 2022). Therefore, the economic status should be taken
into consideration when formulating welfare policies.
The method of delivery also affects the impact of education subsidy schemes. Results indicated that the means-tested schemes with flat rates had a higher adequacy, in terms of the amount of benefits reaching poor households, than those with sliding scales (Maggie, et al., 2019). It is also found that universal benefits have more substantial impacts on all child poverty and income inequality indicators compared to current means-tested benefits in Hong Kong due to potential exclusion errors (Maggie, et al., 2019).
- Methodology
Hendren & Sprung-Keyser (2020) used the MPVF approach to evaluate the impact on social welfare of all social welfare programs in the US. The equation of MPVF is shown below.
However, the limitation of this approach is that utility is subjective, especially between the rich and the poor. Therefore, the methodology of this paper is a modified version of the approach used in Hendren & Sprung-Keyser (2020). Similar to Hendren & Sprung-Keyser (2020), the net cost combines both direct spending on the transfer payment and the long-run impact of the policy on the government’s budget (i.e. fiscal externalities). As quality education and future
income are shown to have a positive relationship (Chenhong, et al., 2019), educational subsidies enable students from humble family backgrounds to have an opportunity to climb the social ladder. This increase in future income will increase government tax revenue, which could offset the costs of the subsidy scheme. On the other hand, the benefit is calculated by combining the reduction of educational expenses for families and the increase in average monthly income in the future. I then take the ratio of the benefits to net government costs, which gives us the increase in monthly income for individuals from low-income families per dollar of government spending on the policy.
I use this adopted methodology from Hendren & Sprung-Keyser (2020) since the aim of the educational subsidy scheme is equalizing the opportunity to increase future earnings through education. It would be more appropriate to measure benefit as the increase in future income than the willingness to pay. Administrative costs such as monitoring costs and expenses related to verifying applicant eligibility are not calculated as detailed breakdowns of administrative costs are not always readily available or publicly disclosed. Furthermore, these costs are often relatively small compared to the actual disbursement amounts. Their impact on the overall benefit-to-cost ratio might be minimal. I also acknowledge that there are non-monetary benefits such as increased productivity, that are difficult to quantify. Therefore, these benefits are not included in the calculation. This paper uses a straightforward methodology that focuses on the most significant factors related to reducing income inequality.
The average amount of subsidy disbursed data in 2022-23 provided by the Working Family and Student Financial Assistance Agency(WFSFAA) is used to calculate the costs of subsidy schemes.
For post-secondary subsidy schemes, I use the median income of degree-holding workers as a proxy for the impact of these subsidies on future earnings. Through analyzing cross-sectional data from the Census and Statistics Department of Hong Kong (Census and Statistics Department, 2024), I compare income disparities between subsidy recipients and non-recipients, estimating the benefit component of the analysis. I measured average income using median income level to reduce the impact of very high earnings outliers.
2.1 Comparison with Other Methodologies
Regression analysis to analyse the drop in the Gini coefficient after the implementation of a social welfare program is often used to evaluate the effectiveness of welfare policies (Abdullah, et al., 2013; Mehr, et al., 2024). Some also use descriptive data from questionnaires to measure the satisfaction of citizens towards those programs (Mehr, et al., 2024). However, these methods might not be most suitable for evaluating the effectiveness of education subsidy schemes. As compared to regression analysis, the cost-benefit analysis used in this study accounts for future outcomes and delayed effects of the education subsidy schemes, which regression models might struggle to capture accurately. This cost-benefit analysis method incorporates both indirect costs and long-term fiscal impacts into the calculation, capturing the indirect effects of education subsidy schemes, whereas regression analysis can only capture the immediate effects of such schemes. Also, showing the change in income inequality using the Gini coefficient has its limitations. The Gini coefficient does not capture the nuances of income distribution effectively. It can remain unchanged even if the income of the poorest or richest individuals changes, as long as the overall distribution remains the same. It is less effective in capturing changes in inequality
within specific segments of the population and the between-group inequalities. As for qualitative analysis, it fails to quantify the return on investment for government spending. The cost-benefit analysis used in this study provides a standardized metric (income increase per dollar spent) for comparing different policies.
- Result
Subsidy schemes are categorized into two different groups, namely, 1) primary & secondary level and 2) post-secondary & tertiary level. I will segment the results into these two sections. The cost-benefit ratio of each scheme is listed in Table 1.
3.1 Primary & Secondary Level
Schemes targeting primary & secondary level include School Textbook Assistance, Subsidy Scheme for Internet Access Charges, Student Travel Subsidy Scheme, Examination Fee Remission Scheme, Student Grant, Financial Assistance Scheme for Designated Evening Adult Education Courses (FAEAEC) and Diploma of Applied Education (DAE) / Diploma Yi Jin (DYJ) Tuition Fee Reimbursement. The first five subsidy schemes are nondistortionary transfers from the government to individuals. Since the benefits of these schemes are the reduction in the individuals' education-related expenditure, the cost-benefit ratio is 1. It is important to note that the Student Grant is a universal benefits subsidy scheme, which differs from the 4 other means-tested schemes.
FAEAEC has an average amount of subsidy disbursed of HKD$5,525 in the 2022-23 period. FAEAEC will provide subsidies for people aged 17 or above receiving secondary-level education. It is created for people in the labour force who have not attained lower or upper secondary-level education. In 2023, 3.6% of the employed population has an education attainment level of lower secondary education (C&SD, 2023), which means that they could apply for upper secondary education under the FAEAEC scheme. The monthly wage gap between workers with lower secondary education and those with upper secondary education is HKD$2,800. After three years, which is the typical duration of upper secondary education, this HKD$2,800 increase in income has a net present value of 2800/(1+4%)^3 = HKD$2,489, given the social discount rate in Hong Kong is 4%. As workers can continue their employment while attending evening school, there is no income forgone. Therefore, the benefit level is HKD$8014. As the increase in wages will yield a HKD$3.33 increase in income tax revenue per month per person for the government, the net cost will be $5521.7. This gives us a benefit/cost ratio: (5525+2,489)/95525-3.3) = 1.45. For the Diploma of Applied Education (DAE) / Diploma Yi Jin (DYJ) Tuition Fee Reimbursement, the calculation is similar. The average amount of subsidy disbursed is HKD$11,120. The calculation method is the same.
It is important to note that FAEAEC is a conditional cash transfer scheme while DAE/DYJ Tuition Fee Reimbursement is an unconditional cash transfer. The difference between these two transfer modes, namely the tradeoff between administrative costs and positive externalities on society, will be discussed in later sections.
3.2 Post-secondary & Tertiary Level
The education subsidy schemes targeting post-secondary & tertiary level students include the Financial Assistance Scheme for Post-secondary Students (FASP), and the Tertiary Student Finance Scheme - Publicly-funded Programmes (TSFS).
First, for FASP, 46.8% of the students applying are pursuing a sub-degree and 53.2% are pursuing a degree. According to WFSFAA data, 12476 students receive average subsidies of HKD$58822 in the form of grants. As FASP requires applicants to be full-time students, applicants have to forgo 4 years (2 years for sub-degree students) of potential income, which will be deducted from the benefits. Students are looking for higher future income at the expense of the potential immediate income forgone. The NPV of the increase in future income is calculated for the first 8 years after graduation. Beyond that timeframe, the income projections would be affected by other variables such as additional vocational training, further education, job changes, and other professional development opportunities. I also take into account the tax revenue foregone from students who delay entering the workforce and the increase in future tax revenue from these students after their graduation. For TSFS, all applicants are pursuing a degree. The calculation method is the same as above. Just by comparing the benefit/cost ratio, TSFS has more impact on welfare than FASP, which is obvious because obtaining a university degree has a larger impact on increasing their future income than a sub-degree. The purpose of subsidizing students pursuing sub-degree
programs is to encourage them to subsequently enrol in and complete a full university degree, expecting that this will lead them to ultimately attain a university-level qualification.
3.3 Comparison with Comprehensive Social Security Assistance (CSSA) Scheme
CSSA Scheme is one of the largest welfare schemes in Hong Kong. However, this scheme has a benefit/cost ratio of less than one. First, the cost is more than just payment amounts, including huge administrative costs as applicants need to pass financial tests and are selected by Social Welfare Department workers. Moreover, there is no future increase in government expenditure as it doesn’t increase the future income of the recipient as an education subsidy scheme does. It might even decrease government revenue as these cash transfers might disincentivize workers from earning income, further increasing the burden on the government budget. Therefore, it is clear that education subsidy schemes are more cost-effective than the CSSA Scheme.
- Discussion and Limitations
4.1 Discussion
The result shows that investing in subsidy schemes targeting university students has the largest impact on increasing future income per dollar invested. There are two points worth discussing.
First, the choice between a means-tested scheme and a non-means-tested scheme. Non-means-tested educational schemes provide opportunities to all students regardless of their financial background (Chaitra, et al., 2024) – means-tested schemes are especially unfair for families who are just above the income eligibility threshold. Also, adopting non-means-tested schemes reduces the administrative costs of the government while improving social mobility. Therefore, inclining towards universal benefits schemes like the $2500 Student Grant for secondary school students would be optimal for the government.
Second, the choice between conditional cash transfers (CCTs) and unconditional cash transfers (UCTs). A CCT makes its payment conditional on the completion of a behaviour, and a CCT will have a greater positive welfare impact than a UCT if the encouraged behaviour has a positive externality, in which the social benefit is greater than the social cost, for instance, an education subsidy scheme. However, unconditional cash transfers (UCTs) are cheaper to deliver and administer because no monitoring of conditions is required. This leads to a fundamental tradeoff that policymakers designing transfer programs must grapple with – the tradeoff between the positive externality of the welfare programme and the administrative cost.
Although CT programmes indicate increased school attendance in both CCTs and UCTs, there was no significant difference between CCTs and UCTs programmes in terms of school attendance (Baird et al., 2014). However, there were positive links found between access to the CTs and the chances of school attendance and enrolment compared to when there is no CT in place. In terms of education, institutional factors such as the capacity of teachers, parental support and overall school environment play a more crucial role in providing improved learning outcomes (Samson et al., 2010). This leads us back to the point that was discussed earlier in the result section – the ultimate goal of education subsidy schemes should be to increase students’ access to university education. For example, the purpose of subsidizing students pursuing sub-degree programs is to encourage them to subsequently enrol in and complete a full university degree, expecting that this will lead them to ultimately attain a university-level qualification. Therefore, it is important to enhance students’ academic proficiency by increasing primary and secondary education quality for underprivileged students through subsidy schemes, equipping students with the ability to receive university education. Therefore, while CCTs might be more suitable for sub-degree subsidy schemes (Primary and secondary education in Hong Kong is already compulsory. Therefore, implementing CCT schemes for secondary school subsidies would likely be redundant and inefficient.), UCTs are more efficient than CCTs in university subsidy schemes. There is no need to set conditions for cash transfers as there is no under-investment in this socially desirable behaviour. When markets operate without friction, UCTs will dominate CCTs on efficiency grounds, since imposing a condition or constraint can only make the beneficiary worse off and increase administrative costs.
Some research found a declining trend in college wage premiums over time, suggesting a cumulative disadvantage in terms of earning potential for those born after 1950 (Xiaogang, et al., 2022). Individuals holding a university degree in Hong Kong do not always experience an earnings advantage, challenging the assumption of a direct correlation between education and income in the financial sector. This challenges the belief of investing more in educational subsidy schemes. However, the cause of this phenomenon is not that university education is useless, but rather the relatively high prevalence of education–occupation mismatch in Hong Kong, especially in the finance sector (Chi, et al., 2020). The job-specific skills and knowledge required can usually be acquired through on-the-job professional development activities offered by employers and professional development programmes. Therefore, simply holding a university degree does not give a huge competitive edge to students. The real advantage of university education is the soft skills gained during the process, such as communication skills, problem solving skills and creativity. In addition, a university degree increases post-school investment in human capital, for instance, job training and career investment, which ultimately leads to an income premium (Haodong, et al., 2024). Huggett et al. (2011) found that differences in initial human capital and learning ability at age 23 explained a 61.5% variation in lifetime earnings. Focusing on developing the soft skills of students will help increase their future income. Education is believed to increase human capital or the productivity of workers, which would increase their earnings. Also, the theory of education signalling suggests that under asymmetric information in the labour market, employers generally believe that more educated employees have higher levels of productivity, which should earn higher wages. Thus education increases the signal to increase individual income (Manxue, et al., 2016). Therefore, it is still worth investing in education subsidy schemes to reduce income inequality.
4.2 Limitation
I assumed that the average income data remained relatively constant throughout the years that the students were in university. This slightly undermined the accuracy of the calculation. Track changes in income over time is crucial for capturing the dynamic effects of subsidy programs on income distribution. Future studies could track a cohort of students under the education subsidy scheme and a control group over an extended period, then use analysis methods such as Difference-in-differences to more precisely examine the impact of the subsidy scheme on participants' future income. Also, students with a university degree might have a greater promotion prospect than those with no degree. This is not accounted for in the calculation.
- Conclusion
This study has examined the effectiveness of various educational subsidy schemes in Hong Kong in reducing income inequality. The analysis reveals that these schemes, particularly those targeting post-secondary and tertiary education, have a positive impact on reducing income disparities and offer a favourable benefit-to-cost ratio. Tertiary education subsidies, such as the Tertiary Student Finance Scheme (TSFS), demonstrate the highest benefit-to-cost ratio (2.073), indicating their significant potential in enhancing future earnings and reducing income inequality. Educational subsidies outperform other welfare schemes, such as the Comprehensive Social Security Assistance (CSSA) Scheme, in terms of cost-effectiveness and long-term impact on income.
In addition, the Hong Kong government should prioritize universal benefit schemes, like the Student Grant for secondary school students, as they may be more effective than means-tested
schemes in improving social mobility and reducing administrative costs. Also, unconditional cash transfers appear more suitable for Hong Kong's context, given the cultural emphasis on education and the high motivation levels of students from low-income backgrounds. It is important to note that these subsidy schemes should be refined according to the demand in the labour market to enhance the effectiveness of these programs in reducing income inequality.
In conclusion, while educational subsidy schemes alone may not completely solve the issue of income inequality in Hong Kong, they represent a crucial and effective tool in the government's arsenal for promoting social mobility and creating a more equitable society. Continued investment and refinement of these programs, coupled with broader economic and social policies, will be essential in addressing the persistent challenge of income disparity in Hong Kong.
References
Abdullah, A., Doucouliagos, H., & Manning, E. (2013). DOES EDUCATION REDUCE INCOME INEQUALITY? A META-REGRESSION ANALYSIS. Journal of Economic Surveys, 29(2), 301–316. doi:10.1111/joes.12056
Ahmed, Raza, Cheema., Shanza, Ghaffar., Jabbar, Ul-Haq., Qazi, Muhammad, Adnan, Hye. (2023). Does educational inequality lead to income inequality? Evidence from Pakistan. Edelweiss applied science and technology, 7(2):154-163. doi: 10.55214/25768484.v7i2.404
Baird, S., Ferreira, F. H. G., Özler, B., & Woolcock, M. (2014). Conditional, unconditional and everything in between: a systematic review of the effects of cash transfer programmes on schooling outcomes. Journal of Development Effectiveness, 6(1), 1–43. https://doi.org/10.1080/19439342.2014.890362
Bryan, G., Chowdhury, S., Mobarak, A. M., Morten, M., & Smits, J. (2023). Encouragement and distortionary effects of conditional cash transfers. Journal of Public Economics, 228, 105004.
Census and Statistics Department. (2024). Table 220-23013 : Monthly wage level and distribution by employment nature and educational attainment. Web table.
https://www.censtatd.gov.hk/en/web_table.html?id=220-23013
Chaitra, Rangappa, Beerannavar., Shainy, Pancrasius. (2024). National Education Policy 2020. Advances in educational marketing, administration, and leadership book series, 138-158. doi: 10.4018/979-8-3693-1614-6.ch008
Chenhong, Peng., Paul, S., F., Yip., Yik, Wa, Law. (2019). Intergenerational Earnings Mobility and Returns to Education in Hong Kong: A Developed Society with High Economic Inequality. Social Indicators Research, doi: 10.1007/S11205-018-1968-2
Chetty, R., Friedman, J. N., Hilger, N., Saez, E., Schanzenbach, D. W., & Yagan, D. (2011). How Does Your Kindergarten Classroom Affect Your Earnings? Evidence from Project Star. The Quarterly Journal of Economics, 126(4), 1593–1660. doi:10.1093/qje/qjr041
Chi, Wai, Chan. (2020). The earnings advantage of university education: a case study of the financial sector in Hong Kong. Educational Research for Policy and Practice, doi: 10.1007/S10671-018-9241-7
Encouragement and distortionary effects of conditional cash transfers. IZA. (2021). https://www.iza.org/publications/dp/14326/encouragement-and-distortionary-effects-of-conditional-cas h-transfers
Haodong, Qi., Martin, Kolk. (2024). How Does University Education Pay Off? – A Time-Series Analysis of Individual Life-Cycle Earnings. doi: 10.21203/rs.3.rs-3866234/v1
Hendren, N., & Sprung-Keyser, B. (2020). A Unified Welfare Analysis of Government Policies*. The Quarterly Journal of Economics. doi:10.1093/qje/qjaa006
Huggett, M., G. Ventura, and A. Yaron (2011, December). Sources of Lifetime Inequality. American Economic Review 101 (7), 2923–54.
Maggie, Lau., Kee, Lee, Chou. (2019). (1) Targeting, Universalism and Child Poverty in Hong Kong. Child Indicators Research, doi: 10.1007/S12187-018-9540-9
Manxue, Chen., Kangyin, Lu., Chengcheng, Tao. (2016). Literature Review on the Impact of Education on Individual Income. 468-471. doi: 10.2991/ICSSHE-16.2016.118
Mehr, A. S., Abdellahi, E., Gandomani, S. J., & Ghahar, S. H. (2024). Analysis of The Effects of Government Policies on Reducing Social and Economic Inequalities. International Journal of Applied Research in Management, Economics and Accounting, 1(2), 69-79.
Minhui, Liu., Lok, Sang, Ho., Kai, Wai, Huang. (2022). Upward Earnings Mobility in Hong Kong: Policy Implications Based on a Census Data Narrative. The China Quarterly, doi: 10.1017/S0305741022001230
Socioeconomic indicators - Hong Kong: Market forecast. Statista. (2024, May).
https://www.statista.com/outlook/co/socioeconomic-indicators/hong-kong
Student Finance. Working Family and Student Financial Assistance Agency. (n.d.-a). https://www.wfsfaa.gov.hk/en/sfo/index.htm
Table 210-06304 : Employed persons by educational attainment, age and sex. (2023). https://www.censtatd.gov.hk/en/web_table.html?id=210-06304
Vu, Khanh, Quy. (2022). Income and Wealth Inequality in Hong Kong, 1981–2020: The Rise of Pluto-Communism?. The World Bank Economic Review, doi: 10.1093/wber/lhac019
Xiaogang, Wu., Maocan, Guo. (2022). Higher education expansion and the changing college wage premium in Hong Kong, 1976–2016. Chinese journal of sociology, 8(1):3-28. doi: 10.1177/2057150x211073453
Zhang, P. (2022). Can public subsidy on education reduce wage inequality in the presence of automation?. Economic research-Ekonomska istraživanja, 35(1), 6850-6866.
