Journal Issue

People’s Republic of China—Hong Kong Special Administrative Region Selected Issues

International Monetary Fund. Asia and Pacific Dept
Published Date:
January 2019
  • ShareShare
Show Summary Details

Income Inequality in Hong Kong SAR1

Income inequality in Hong Kong SAR remains high, despite declining recently. Redistributive policies implemented by the authorities have helped to lower income inequality. But inequality is likely to rise in the medium-term due to aging and thus more needs to be done. A package of policies could lower the Gini index by 3–4 points by 2050 including: more progressive salaries tax; higher reliance on recurrent property taxes; and increased public expenditure on social welfare, health, housing, education and child care. According to recent evidence in the literature, these policies could also boost growth by 0.2–0.5 percentage points per year.

A. Background

1. Income inequality in Hong Kong SAR remains high, both historically but also compared to other economies. The market Gini coefficient, which measures household income inequality prior to taxes and transfers, stood at 49.9 in 2016, while the net Gini coefficient, reflecting the impact of redistributive policies in the form of taxes and transfers, stood at 42. Both coefficients declined compared to the 2011 level, indicating a small reduction in income inequality. Comparison with other economies indicates that inequality in Hong Kong SAR is relatively high, even when measured post taxes and transfers, and it appears higher than in most other cities with large financial centers. That said, the data shows that the increase since 1990 has not been as high as in Mainland China, the U.S. or other advanced economies, both in Asia and elsewhere.23

Evolution of Gini Coefficients in Hong Kong SAR 1/

(Per Capita Gini; Index)

Source: SWIID 7.1; Hong Kong SAR C&SD; and IMF staff calculations.

Note: 1/ Inequality measured by Gini coefficients. Shaded area represents the degree of redistribution.

Levels and Trends in Income Inequality Across Countries

(Per Capita Gini; Net Gini Coefficient)

Sources: SWIID 7.1 and IMF staff estimates.

Gini coefficients and degree of redistribution

(Per Capita Gini; Percentage points)

Sources: SWIID 7.1 and IMF staff estimates.

Market Gini Index: Hong Kong SAR and Other Cities with Sizable Financial Industry

(Household Gini; Percentage points)

Sources: U.S. Census, Statistics Singapore, HKSAR C&SD, OECD.

1/ Southern Kanto region; 2/ Singapore’s Gini coefficient is reported based on household income from work per household member.

2. Government policies are helping counteract the effects of rising inequality. Between 1996 and 2016, market per capita Gini rose by 0.6 points but the net Gini actually declined by 0.6 points, indicating a rising degree of redistribution due to government policies, measured as the difference between the two Gini indices. A similar picture emerges when looking at the long-term trends in the Standardized World Income Inequality Database (SWIID): between 1964 and 2016, the degree of redistribution increased from 3.4 Gini points to 4.6 Gini points. Redistribution in Hong Kong SAR is also higher than in many other economies in Asia, though significantly below levels observed in other advanced economies.

3. The authorities put in place policies to help those at the bottom of the income distribution. Progressive taxation rates, a statutory minimum wage introduced in 2011 and increased since, various family and old age allowances, which have been reformed in 2017 and 2018 to further enhance the retirement protection system and social security coverage of working families, transportation subsidies, as well as housing, education, and health benefits, helped counteract income inequality. These policies in total appear to be progressive: when comparing household income post taxes and social transfers, households in the lowest decile of the income distribution receive close to 125 percent of their original income in additional transfers, while net income of households in the highest decile decreases by around 10 percent.

4. …and rising incomes have helped. Since 2006, the average real median household income increased on average by almost 24 percent in the lowest two income deciles. That increase is almost twice as high as for the rest of Hong Kong SAR’s population, where the cumulative real growth—while still significant—did not exceed 10 percent.

Household post-tax and transfer income to original income

(Percent per income decile)

Source: Hong Kong SAR 2016 Household Survey.

Average growth rate of real household income 1/


Sources: Table 2.3, 2016 Hong Kong SAR Household Survey and IMF staff estimates.

Note: 1/Median monthly income from main employment at constant prices by decile group of working population.

Domestic housing affordability indicators

(Q1 2004 = 100)

Sources: Haver and IMF staff estimates.

Distribution of benefits by household income quintile


Sources: Hong Kong SAR C&SD and IMF staff estimates.

5. …but problems remain. Levels of inequality in Hong Kong SAR remain high, both by historical standards and by international comparison. Over 10 percent of the population lives in poverty (with the poverty line defined as half of the median income for a given household size)—even after the policy measures—and this ratio has increased since 2014. Sharply rising house prices, which outpaced wage growth, put private housing out of reach for a significant share of the population; consequently, around 30 percent of all households live in public rental housing, where the average wait time has increased from 1.8 years for family applicants and 1.1 years for the elderly in 2008, to 5.5 years and 2.9 years, respectively, in September 2018, and additional 15 percent of households live in subsidized sale flats. The share of housing benefits going to households in the lowest income quintile has been falling while the share of government spending dedicated to housing has remained steady, at around 1.1 percent of GDP since 2005.

Poverty rates


Sources: Hong Kong SAR C&SD and IMF staff estimates.

6. Looking at equality of opportunities, Hong Kong SAR compares well on many of the indicators, though pockets of inequality remain. Over 95 percent of the population have access to financial services; adult literacy rates, as well as secondary and primary school enrollment are very high and similar to other advanced economies, while on PISA scores Hong Kong SAR surpasses its peers. However, achievement differentials do exist between children based on their family’s income, and between children who do and do not speak Chinese at home (with non-Chinese speaking children considered to be at a disadvantage), indicating that more resources might be needed to address these challenges.

Education Indicators

(Latest value available)

Source: IMF FAD Expenditure Assessment Tool (EAT), World Bank.

Gaps in Learning Achievement 1/

(TIMSS 2015, 4th level, difference in percentage points)

Sources: World Education Inequality Database; and IMF staff estimates.

Note: 1/ Wealth gap defined as a difference between households in the bottom 20 percent vs. top 20 percent of income distribution. Language gap: whether child speaks Chinese at home.

7. Levels of pension and health coverage are also high, but there is room to increase the safety net for the elderly. While health coverage extends to the whole population and health outcomes in Hong Kong SAR compare favorably to other advanced economies, the share of out-of-pocket medical expenses is higher than in other advanced economies. Among the elderly, the share receiving old-age pension is lower than in other advanced economies (though it is important to note that over 70 percent of Hong Kong SAR’s population aged 65 and above receive social security, and the share rises to 87 percent for those over 70) and close to 40 percent of elderly households live in poverty.

Health and Health System Characteristics Indicators

(Latest value available)

Source: IMF FAD Expenditure Assessment Tool (EAT), World Bank, World Health Organization.

Poverty rates among elderly households


Sources: Hong Kong SAR C&SD and IMF staff estimates.

Health and Health System Characteristics Indicators

(Latest value available)

Source: IMF FAD Expenditure Assessment Tool (EAT), World Bank, World Health Organization.

Share of population receiving old-age pension


Source: ILOSTAT.

B. Empirical Analysis

In the years to come, aging is projected to push inequality up. More active government policies, including increased tax progressivity higher public spending on transfers, health and social welfare, and public housing provision could reduce the Gini index by 3½ points by 2050. At the same time, growth does not need to be lower. Based on the empirical relationship between growth and inequality established in the literature, growth could be boosted by 0.2–0.5 percentage points on average per year if these active policies are implemented.

8. In this section, we estimate the link between inequality and its drivers, and project such drivers to create two forward-looking scenarios for inequality in Hong Kong SAR.

Health coverage


Sources: ILOSTAT, and IMF staff estimates.

Share of population receiving old-age pension


Source: ILOSTAT.

Using the same setting as Jain-Chandra and others (2018) (JC et al) and extending the data to include Hong Kong SAR, we first document the main drivers of inequality in 1992–2010 splitting them into structural and policy variables. We then formulate a baseline scenario in which structural variables are projected forward and policies respond only passively. In addition, an “active policy” scenario is formulated and its impact on inequality assessed. Finally, estimates from the literature on the link between inequality and growth are used to generate a range of impacts of such “active policies” on growth.

9. We estimate the drivers of inequality using the approach in Jain-Chandra and others, 2018. Concretely, we estimate a fixed-effects panel regression that spans 1980–2010 and 29 economies, including Hong Kong SAR. The inclusion of Hong Kong SAR is the only difference to the analysis in JC et al.

10. Drivers can be divided into two groups: structural factors and policies.4

  • Structural factors: (i) urbanization measured by the share of the population living in urban areas; (ii) aging as represented by the Higgins (1998) polynomial; (iii) sectoral change as measured by the share of employment in the services and industry sector; and (iv) educational levels measured by the share of the population with higher education.
  • Policies: (i) individual income tax revenue; (ii) property tax revenue; (iii) public spending on health; (iv) public spending on social protection; and (v) overall redistribution as proxied by the difference between the Gini index before (market Gini) and after transfers and taxes (net Gini). Policies (i)-(iv) are measured as a share of GDP. The inclusion of the two types of taxes is due to their progressive nature, either relying on progressive scales (individual income tax more often than property taxes) or falling disproportionately on high-income/wealth households.

11. Including Hong Kong SAR in the empirical analysis retains all of the main conclusions of JC et al (see Table 1).5 Given that HKSAR is an outlier in many dimensions (being a major financial center) and is often hard to treat in a panel setting, the fact that results are broadly consistent between JC et al (column 1 of Table 1) and those here (column 2 of Table 1) is reassuring. There are some differences though that are worth outlining. Estimates for structural variables are very comparable, except those pertaining to the industry share in employment which are somewhat weaker but still significant at 5 percent level and with the same sign. However, the joint impact of the two industry share variables (linear and quadratic terms) are similar across the two specifications, because both terms are smaller in absolute terms in column (2) compared to (1) while still having opposite signs. Employment in services was barely significant in (1) and is now insignificant. Results for structural variables are intuitive and consistent with the literature. Within the policy variables, the main difference is a slightly higher coefficient on individual income tax which is now marginally significant. Other policies are very comparable across columns (1) and (2). In general, higher spending on health and social protection, more redistribution and higher tax revenues on properties and individual income taxes all lower inequality as measured by the Gini after transfers and taxes.

12. Having established an empirical link between inequality and its drivers, we project structural variables and passive policies forward to forecast inequality under a “Baseline Scenario”. Structural variables are easier to predict. Population trends are taken from the UN’s medium-variant population projections, urbanization and sectoral variables are held constant given full urbanization and preponderance of services in Hong Kong SAR. Educational attainment is also held constant6 As for policies, two are held constant for lack of specific policy intentions in these areas (redistribution and individual income tax), but the other three are allowed to adjust passively to projected changes in conditions, most importantly demographics. For the two expenditure variables, we use the projections in Mano (2017) that build on Clements and others (2015), Clements and others (2013) and IMF (2017), and essentially assume that the demographic structure drives the path of public expenditures on health and social protection. Property tax revenues are assumed to decline to the level consistent with a zero house price gap by 2030, analogous to analysis in Mano (2017) applied to property tax revenues rather than to headline fiscal revenues.7

Net Gini: Hong Kong SAR

(Baseline versus Active Policies Scenarios)

Sources: SWIID 7.1 and IMF Staff Estimates.

13. Under the Baseline Scenario, inequality is expected to rise further. Inequality declined by 0.7 Gini points between 1996 and 2016. Taking our baseline projections for structural variables and policies, the net Gini could rise by 2.6 points between 2016 and 2050. This underlines the challenge of tackling inequality in Hong Kong SAR. Note that inequality rises despite our assumption that public expenditure adjusts naturally to demographic trends, something that alleviates some of the inequality generated by rapid aging.

14. An “active policies” scenario is then constructed. In this scenario, structural variables are kept unchanged from the baseline scenario, but policies are adjusted to combat rising inequality. In particular, all policies except property tax revenue are assumed to converge to the mean of those in other financial centers8 by 2050. This would entail a rise of 7.7 percentage points of GDP in public social expenditure, of 0.7 percentage points of GDP in public health expenditure, of 8.8 points in the coefficient of absolute redistribution, and of 6.2 percentage points of GDP of individual income tax revenue. For property tax revenue, Hong Kong SAR’s revenues under the baseline are projected to be higher than the current average of other financial centers, this despite the projected decline due to the normalization of the housing cycle. Because of its economic structure, Hong Kong SAR is likely to continue to rely on property taxation over the long run and thus the active policies scenario assumes that recurrent property taxes are raised to partially offset the loss of revenue with the eventual removal of stamp duties in the housing market as the house price gap closes.9

Baseline and Alternative Scenarios

(in percent of GDP1)

Sources: IMF Staff Estimates.

Note: Except for absolute redistribution.

Contributions (with baseline policies)

(Gini points)

Sources: IMF Staff Estimates.

15. Under “active policies”, inequality declines by 0.8 points from its 2016 level. The active policies package prevents inequality from rising 2.6 points, and thus its effect is to lower Hong Kong SAR’s net Gini by 3.4 points by 2050 relative to the baseline. It is instructive to decompose the changes in net Gini in both baseline and active policies scenarios.

16. Structural factors have contributed to a rise in inequality and are expected to continue to pose a challenge, particularly aging. In the past, the move from industry to services created some inequality and aging was not a major factor as the demographic transition was still in its inception. However, Hong Kong SAR’s population is expected to age quickly in the medium-term and this is by far the largest challenge to inequality, possibly pushing the net Gini index higher by close to 5 points by 2050.

Contribution of Structural Trends

(Gini points)

Sources: IMF Staff Estimates.

17. Policies have been supportive in the past, but if not adjusted, will lose some of their redistributive power. In the past, larger redistribution and property taxes have been successful in containing increases in inequality due to structural change. But going forward, under the baseline scenario, policies will not be able to further contain rising inequality due to aging. Health expenditure is assumed to increase in line with changing demographics and thus it is not surprising that it is the one policy that seems most effective under the baseline.

Contribution of Policies

(Gini points)

Sources: IMF Staff Estimates.

18. Under the active policies scenario, all policies make a significant contribution to reducing inequality. After the needed rise in health expenditure, the largest impact is due to higher overall redistribution, whose contribution to lowering inequality is 1.6 Gini points compared to the baseline. This variable is a summary statistic of all efforts to contain market-generated inequality and is shown to be crucial in preventing the rise of inequality in the baseline. It would include all efforts mentioned below plus those to expand public housing supply among others. Public social protection expenditure comes second contributing 0.9 points to lowering the Gini compared to baseline. This would entail strengthening programs, like Working Family Allowance, Old Age Living Allowance, child care policies and boosting public education. Higher and more progressive individual income taxes would contribute another 0.7 points. Property taxes which were declining significantly under the baseline, contribute to lowering the Gini index by 0.1 points compared to baseline with a switch to higher reliance on recurrent property taxes which are less dependent on the housing cycle. Health expenditure is not much changed between baseline and active policies scenarios reflecting the strong increase that it is already embedded in the baseline.

19. Thus, active policies generate a decline of the net Gini of 3.4 points. But what about the impact on growth?

20. Although the early literature found a positive relationship between initial inequality and subsequent growth, several recent studies found the opposite. Two papers utilized the Deininger and Squire Database, 1996, where income inequality is measured by the Gini index after transfers and taxes. Barro (1999) found that the effect of inequality on growth was negative for per capita GDP below around $5000 (in 1985 USD) and positive above, implying that increases in income inequality negatively affect growth in poor countries, and positively in rich countries.10Forbes (2000) found a significant positive link between inequality and growth in the short- and medium-term. More recently, however, several studies have found a negative relationship between inequality and growth (Castello-Climent (2007), Cingano (2014), Ostry et al (2014), and Dabla-Norris and others (2015)). Table 2 summarizes these results, focusing on estimates of how the Gini index affects growth.

21. There are various channels through which higher inequality could lead to lower growth.

22. Reducing income inequality could raise growth in the medium term. Applying the literature’s more recent estimates, the reduction of 3.4 points in the net Gini envisioned in the active policies scenario could result in higher growth of real per capita GDP by 0.2 to 0.5 percentage points per year compared to the baseline scenario. Given that Hong Kong SAR is an advanced economy, its growth rate is expected to stay in lower single digits, and fall over the medium-term due to aging, this would be a significant boost to the city’s potential.

C. Policy Recommendations

More active fiscal policies could be used to lower income inequality without sacrificing medium-term growth prospects.

23. Hong Kong SAR’s tax system could be made more progressive. To that end, the authorities could increase progressivity of the salaries tax, especially at the top, and reverse the recent tax cuts. Recurrent property taxes—general rates and government rent—could be raised. That would lower the reliance on volatile transaction taxes, making the revenue source more immune to the housing cycle. The recurrent property tax schedule could also be made more progressive, with higher taxes levies on high-value properties.

24. Public spending in several areas could be raised to help stem inequality increases.

  • Public spending on social welfare could continue to be raised to boost redistribution and increase access of poorer households.
  • Given the impending aging problem, public health expenditure will need to go hand in hand with the pace of aging, resulting in a considerable expansion of more than a third by between FY2017/18 and 2030, and doubling by 2050.
  • Public social protection expenditure will also need to be expanded, including the Old Age Living Allowance (OALA) and the Working Family Allowance, while targeting could be improved by phasing out Old Age Allowance that is not means-tested in favor of OALA or other means-tested programs.
  • Public housing expenditure should be expanded to help alleviate the acute housing shortage and shorten the average waiting time for public rental housing.
  • The offsetting mechanism in the Mandatory Provident Fund should be abolished, in line with the authorities’ intentions.
  • Spending on education and child care should be raised to help lower the market income inequality directly. The commissioned study aiming to determine the demand and supply for child care services and map out the long-term service development programs, as well as the initiatives mentioned in the 2018 Policy Address, should help in this regard.
  • The level of statutory minimum wage should be revised regularly to keep in line with rising cost of living. The authorities could consider moving to annual from the current biennial reviews if deemed necessary at any stage.
Table 1.Hong Kong SAR: Panel Regression Estimates of the Drivers of Inequality
Dependent variable: Net Gini CoefficientJC et alAdding HKSAR
Structural VariablesShare of Employment in Services-0.320+


Share of Employment in Services Squared0.002


Share of Employment in Industry-0.886**


Share of Employment in Industry Squared0.010**


Age Distribitution D1166.942**


Age Distribitution D21-8.758**


Age Distribitution D310.338**


Share of Population living in Urban Areas1.576**


Share of Population living in Urban Areas Squared-0.011**


Share of Population with Some Tertiary Education-0.160*


Share of Population with Some Tertiary Education Squared0.004**


Policy VariablesPublic Social Protection Expenditure as Share of GDP-0.146**


Public Health Expenditure as Share of GDP-2.314**


Public Health Expenditure as Share of GDP Squared0.183**


Absolute Redistribution-0.138**


Propert Tax Revenue as Share of GDP-0.241


Individual Income Tax Revenue as Share of GDP-0.065


Number of Observations573588
Adjusted R-squared0.9680.967
Country Fixed EffectsYesYes
Year Fixed EffectsNoNo
t statistics in parentheses+ p<0.10, * p<0.05, ** p<0.01
t statistics in parentheses+ p<0.10, * p<0.05, ** p<0.01
Table 2.Hong Kong SAR: Summary Results from Literature on Growth and Inequality
SourceMeasure of inequalityMarginal impact of 1 additional Gini point on average annual real GDP per capitaControls for redistribution?
Dabla-Norris and others, 2015Net Gini index-0.067*No
Ostry et al, 2014Net Gini index-0.144***Yes
Cingano, 2014Net Gini index-0.155**No
Castello-Climent, 2010Income Gini Coefficient-0.053*No
Forbes, 2000Gini Coefficient0.13**No
Barro, 1999Gini Coefficient0.06No
*, **, and *** indicate statistical significance at the 10, 5, and 1 percent levels, respectively.
*, **, and *** indicate statistical significance at the 10, 5, and 1 percent levels, respectively.

    Aghion, P.,E.Caroli, andC.Garcia-Penalosa,1999, “Inequality and Economic Growth: The Perspective of the New Growth Theories,” Journal of Economic Literature, Vol. 37(4), pp. 1615–60.

    • Search Google Scholar
    • Export Citation

    Alesina, A., andR.Perotti,1996, “Income Distribution, Political Instability and Investment,” European Economic Review, Vol. 40(6), pp. 1203–28.

    • Search Google Scholar
    • Export Citation

    Alesina, A., andD.Rodrik,1994, “Distributive Politics and Economic Growth,” Quarterly Journal of Economics, Vol. 109(2), pp. 465–90.

    • Search Google Scholar
    • Export Citation

    Barro, R. J.,2000, “Inequality and Growth in a Panel of Countries,” Journal of Economic Growth, Vol. 5(1), pp. 5–32.

    Bénabou, R.,1996, “Inequality and growth” NBER Working Paper n° 5658, July.

    Bertola, G.,1993, “Factor shares and savings in endogenous growth,” American Economic Review, 83(5):1184–1198.

    Carvalho, L., &Rezai, A.,2015, “Personal income inequality and aggregate demand,” Cambridge Journal of Economics, 40 (2), 491–505.

    • Search Google Scholar
    • Export Citation

    Castello-Climent,2007, “Inequality and Growth in Advanced Economies: An Empirical Investigation,” Institute of International Economics, University of Valencia.

    • Search Google Scholar
    • Export Citation

    Cingano, F.,2014, “Trends in Income Inequality and its Impact on Economic Growth”, OECD Social, Employment and Migration Working Papers, No. 163, OECD Publishing.

    • Search Google Scholar
    • Export Citation

    Claessens, S., &Perotti, E. C.,2007, “Finance and Inequality: Channels and Evidence,” Journal of Comparative Economics, 35 (4): 748–73.

    • Search Google Scholar
    • Export Citation

    Clements, B.,D.Coady,F.Eich, andS.Gupta,A.Kangur,B.Shang andM.Soto,2014, “Equitable and Sustainable Pensions: Challenges and Experience.”

    • Search Google Scholar
    • Export Citation

    Clements, B.,D.Coady,F.Eich, andS.Gupta,A.Kangur,B.Shang andM.Soto,2013, “The Challenge of Public Pension Reform in Advanced and Emerging Economies”, IMF Working Occasional Paper No. 275.

    • Search Google Scholar
    • Export Citation

    Clements, B.,K.Dybczak,V.Gaspar,S.Gupta, andM.Soto,2015, “The Fiscal Consequences of Shrinking Populations,” IMF Staff Discussion Note, SDN/15/21.

    • Search Google Scholar
    • Export Citation

    De la Croix, D., andM.Doepke,2003, “Inequality and Growth: Why Differential Fertility Matters,” American Economic Review, 93 (4), 1091–113.

    • Search Google Scholar
    • Export Citation

    E.Dabla-Norris,K.Kochhar,N.Suphaphiphat,F.Ricka,E.Tsounta,2015, “Causes and Consequences of Income Inequality: A Global Perspective,” IMF Staff Discussion Note, SDN/15/13.

    • Search Google Scholar
    • Export Citation

    Forbes, K. J.,2000, “A Reassessment of the Relationship between Inequality and Growth,” American Economic Review, Vol. 90(4), pp. 869–87.

    • Search Google Scholar
    • Export Citation

    Galor, O. andJ.Zeira,1993, “Income Distribution and Macroeconomics,” Review of Economic Studies, 60, 35–52.

    Galor, O., andO.Moav,2004, “From Physical to Human Capital Accumulation: Inequality and the Process of Development,” Review of Economic Studies, Vol. 71(4), pp. 1001–26.

    • Search Google Scholar
    • Export Citation

    Hong Kong SAR Government, 2016, “Hong Kong 2016 Population By-census – Thematic Report: Household Income Distribution in Hong Kong”.

    • Search Google Scholar
    • Export Citation

    Hong Kong SAR Government, 2017, “The Gini Coefficients of Hong Kong in 2016: Trends and Interpretations”, Half-yearly Economic Report.

    • Search Google Scholar
    • Export Citation

    IMF, 2017, “Fiscal monitor”

    Jain-Chandra, S.,N.Khor,R.Mano,J.Schauer,P.Wingender andJ.Zhuang,2018, “Inequality in China – Trends, Drivers and Policy Remedies,” IMF Working Paper 18/127.

    • Search Google Scholar
    • Export Citation

    Mano, R. C.,2017, “Medium-term fiscal prospects and implications” in People’s Republic of China-Hong Kong Special Administrative Region: Selected Issues

    • Search Google Scholar
    • Export Citation

    Moav, O.,2005, “Cheap Children and the Persistence of Poverty”, Economic Journal, 115, 88–110.

    Ostry, J.,A.Berg,C.Tsangarides,2014, “Redistribution, Inequality, and Growth,” IMF Staff Discussion Note, SDN/14/02.

    Perotti, R.,1996, “Growth, Income Distribution, and Democracy: What the Data Say,” Journal of Economic Growth, Vol. 1(2), pp. 149–87.

    • Search Google Scholar
    • Export Citation

    Persson, T., andG.Tabellini,1994, “Is Inequality Harmful for Growth?” The American Economic Review, Vol. 84(3), pp. 600–21.

    Solt, Frederick,2016, “The Standardized World Income Inequality Database,” Social Science Quarterly97(5):1267–12


Prepared by Emilia Jurzyk, Rui C. Mano, and Ananya Shukla.


See Box 2 in People’s Republic of China—Hong Kong SAR 2016 Article IV Staff Report on discussion of difficulties in comparing Gini coefficients across countries and data sources.


Levels of market and net Gini coefficients used in this paper for Hong Kong SAR differ, as some of the coefficients have been taken from the Hong Kong SAR Census and Statistics Department and from the Standardized World Income Inequality Database on per capita basis, while others are reported on a household level basis (also from the Hong Kong SAR Census and Statistics Department). The narrative, however, remains unchanged.


A thorough discussion of each of these variables and the literature on their connection to inequality can be found in JC et al.


We found significant differences between SWIID based inequality data and official sourced data for HKSAR. We decided to take the latter for this study, while maintaining the dataset in JC et al for all other countries which was based on SWIID data.


Underlying these assumptions is the idea that education, urbanization and sectoral compositions are close to their long-run steady-state levels while ageing is not.


This is done by estimating the historical relationship between the IMF team’s estimates of the house price gap (which is an average over 5 different approaches) and property tax revenues between FY95/96 and FY17/18. Such an analysis finds that the tax revenues in FY17/18 of 4.7 percent of GDP were almost 2 percentage points higher than the level consistent with a zero house price gap (or the full period average since full-sample estimates of house price gaps are around zero).


These include Belgium, Ireland, Luxembourg, Netherlands, Singapore, and Switzerland.


The assumed level of property tax revenue to GDP in 2050 (3.5 percent) is still lower than its current level (4.7 percent), but higher than the current average of financial centers or its “natural” level if the house price gap closes (2.3 and 2.8, respectively).


Refers to growth rate regression with fertility variable omitted (Table 4 in Barro, 1999).


Another older strand of the literature asserts that higher inequality may prompt voters to demand higher taxation and regulation, higher public expenditure programs and transfer payments, which in turn could lower investment and reduce economic efficiency (Bertola, 1993; Alesina and Rodrik, 1994; Persson and Tabellini, 1994; Benabou, 1996; Perotti, 1996) but empirical evidence is weakas recent studies have shown (Ostry, 2014, Cingano, 2014).

Other Resources Citing This Publication