(This essay was published in Hong Kong Economic Journal on 1 January 2014)
Hong Kong’s divorce rate in 2011 of 2.89 persons per 1000 population (excluding domestic helpers) places it among the top 10 highest divorce rate countries in the world. Increases in the divorce rate distort measured household income inequality and exaggerate the rise of static poverty over time. Static poverty is concerned with relative inequality between households at a given point in time. If poverty alleviation policies are formulated only on the basis of the observed distribution of household income at a given time, then it could lead to badly designed policies as the experiences of the US and Europe have shown.
Of even greater importance is the effect of divorce on lowering upward social mobility and in generating dynamic poverty. Dynamic poverty differs from static poverty in that it is about poverty across generations due to the lack of upward mobility in society. This essay discusses how these interrelated phenomena could develop in Hong Kong at an alarming pace.
Divorce and Inequality
Figure 1 and Figure 2 shows the number of divorced men and women per 1000 households by the type of housing tenure they are in. The highest number of divorced persons is found among private housing tenants and is closely followed by those among public housing tenants. The number of divorced persons is much lower among homeowners of subsidized sale flats and even lower among homeowners of private housing units. In the first decade of the 2000s the divorce rate was on average around 50-60% higher among renters than homeowners. The difference in the divorce rate between renters and homeowners was much smaller in the 1980s and 1990s, only around 10-25%.
Figure 1: Number of Divorced/Are Seperated Men per 1000 Households by Housing Tenure
Source: HK Population Census and By‐Census Sample Databases, 1976‐2011
Figure 2: Number of Divorced/Are Separated Women per 1000 Households by Housing Tenure
Source: HK Population Census and By‐Census Sample Databases, 1976‐2011
For most people in Hong Kong, housing wealth is the most important part of household wealth. Housing tenure is therefore a very good proxy for whether a family is richer or poorer. We can conclude from the charts in Figures 1 and 2 that the divorce rate is much higher among poor families than rich ones. It also shows that divorce rates are rising rapidly over time and even more dramatically among the poor.
When the divorce rate is rising, especially when it is rising disproportionately for poorer people than for richer people, then the measurement of household income inequality will become fairly distorted. This is a purely statistical point about the way household income is measured. Below is a stylized example to demonstrate this point.
Suppose there are only two households in society, each consisting of a working husband and wife. One household has income of $40,000 (husband and wife each earning $20,000) and one has income of $20,000 (husband and wife each earning $10,000). Average household income is $30,000 in this two-household society.
Now suppose the husband and wife in the lower-income family get divorced. They each keep their jobs. Now there are three households, one earning $40,000 and two earning $10,000. Average household income has now fallen from $30,000 to $20,000. It has fallen 33% even though everyone’s individual income is unchanged.
This example shows that during a time when individual income is actually unchanging, a rise in the divorce rate among families below the average income is going to pull down the measured rate of average household income. This effect is stronger if the wife of the poorer family was not working before. If she takes a job or receives social welfare support, it is likely for a variety of reasons to be at a lower salary than her husband’s.
Notice in this example that there is also a decrease in the number of persons per household. This has nothing to do with having fewer children or spreading income over a smaller number of people. It is a result of the divorce rate that leads to measured household income being a misleading representation of what is going on in the economy and in society. These statistical effects are not small.
Dynamic Poverty is True Concern
This statistical phenomenon also explains, for example, why the household income gap between the rich and poor is rising much faster than the gap in individual income as shown in my essay published earlier (“Divorce, Inequality, Poverty, and the Vanishing Middle Class (I) – Rekindling Hong Kong’s Magic (Part V)”). In that essay, I found the measured increase in household incomes at the bottom tenth percentile of the income distribution was stagnant after the mid-1990s, but it was rising in the case of individual incomes. In other words, rising household income inequality may be, at least partly, a statistical artifact due to a misreading of the data caused by demographic changes.
Calculations similar to those in this example can also be applied to show why the growth of elderly households and single-person households can statistically produce an increase in the measured inequality of household income. This reminds us once again why it is hazardous to interpret changes in static inequality based on a single number, and thereby inferring there has been a rise in poverty, and even more so to anchor policies on it.
But the rapidly rising number of divorces is suggestive of some very real and alarming concerns about how poverty is being formed in Hong Kong, and how it may affect upward social mobility and cause dynamic poverty across generations. It is conceivable that the wealthy could continue to prosper over generations, while the poor continue to stay behind or even fall further behind.
Of particular concern is the growing number of women who are divorced because remarriage rates are much lower among women than men. In 2011, the remarriage rate of previously divorced, separated and widowed men was 48.1 per 1000 population, but it was only 8.4 among women. Divorced women often have low incomes, try to raise children in broken families, and some become increasingly dependent on the state for support. Rising divorce rates are the product of many factors and some of them can have a profound impact on dynamic poverty.
Factors Determining Divorce
Divorce becomes a sad but necessary solution for unhappy marital partners when the benefits to divorce outweigh the costs to pay. An important driver of rising divorce rates in contemporary society is the ability of divorced women to find alternative support and economic opportunities after marital dissolution. Several factors have shifted the balance in favor of the divorce solution.
First, women have become better educated today relative to men than in the past. This has improved the opportunities for women in the labor force, making them less dependent on men for support and better able to lead a meaningful independent life. So when marriages become unpleasant, women are less willing to tolerate an unhappy situation.
Second, government provision of welfare support for divorced women with low incomes and dependent children has also tilted the balance, making it less costly and more feasible for women to accept divorce. Government assistance is provided with the intention of helping those who are already divorced and in need of support, but inevitably these assistance programs create incentives that tilt the balance for those who are still deciding whether to get divorced.
Third, closely related to the rising education and growing labor market opportunities of women is the declining demand for a larger number of children in the family. Time spent together and raising children are the joint assets of the parents. Getting to a settlement on how best to divide up joint assets is always a complicated problem when marriage breaks down.
With fewer children the amount of joint assets in a marriage declines, especially among poorer households, who often cannot afford to make many investments in their children and in addition have few other jointly held family assets such as owned homes. The net result is that divorce is relatively less costly for marital partners to dissolve their marriage. The propensity to become divorced is thus higher among poor families.
Fourth, the introduction of divorce laws in favor of “no-fault, unilateral” divorces has also increased the divorce rate. To some extent the laws themselves are responding to the social and economic forces that are the primary drivers of the divorce rate. But in accommodating these forces the law at the very least has made it easier for divorce to take place by permitting an unhappy party to exit.
In the US, it has been estimated that “no-fault, unilateral” laws led to a short-run increase in the divorce rate of about 10% for a period of about 10 years. There are many reasons for the limited long term effect in the US. Laws were subsequently changed to stem the tide of divorces. People also adjusted their behavior to prevent unilateral walk-outs from unhappy marriages. In Europe, the introduction of “no-fault, unilateral” divorce increased the divorce rate by about 30%, where the average rate of divorces was 2 per 1000 population in 2002. It appears the impact of these laws is much larger and more permanent in Europe.
In Hong Kong, following the recommendation of a 1992 report by the Law Reform Commission, the Matrimonial Causes Ordinance was amended in 1997 to permit divorce after a one-year period of continuous separation, down from three years previously, provided there was proof of an irretrievable breakdown of the marriage and, hence, legitimate grounds for divorce, and provided both parties consented to the divorce. The period of separation was set at two years if only one party consented. This change, while formally preserving the “proof of fault” nature of divorce, effectively made it closer to a “no-fault, unilateral” law.
Fifth, the divorce rate in Hong Kong accelerated upward from the mid-1990s and rose to 2.89 per 1000 population in 2011, placing it in the top 10 in the world and ahead of Japan, China and Europe at 2.0 and Singapore at 1.5, but behind the US at 3.6. What explains this? I conjecture a unique driver of the rapid rise of divorce rates in Hong Kong is the public rental housing program.
Hidden Incentives for Divorce
Hong Kong has operated a public rental housing program since the 1950s. But since the 1980s private property prices have been rising relentlessly as a result of demand growth due to economic globalization and China’s opening, and the slow process of making land conversions. Homeowners experienced an enormous capital gain as property wealth rose. Public rental housing tenants could not partake in this considerable appreciation of wealth. These families literally became the “have-nots” of Hong Kong society.
The deterioration in their relative wealth position, and sometimes also their absolute wealth position as inflation took a toll, was immense. Most of this drop was unanticipated and therefore likely to create considerable marital stress for families who were “have-nots”, including sometimes prospective families in the making. The original marriage contract became sub-optimal for some families. During this period more women also joined the labor force while fewer men did. Divorce appeared less fearful and more attractive to some women.
The public rental housing program had an in-built incentive that provided encouragement for unhappy couples to divorce. After divorce, they became single divorced parents with dependent children. One parent would be able to remain in the public rental housing unit, while the other divorced parent ended up renting housing in the private rental market. This created considerable demand for private housing where supply was limited. Sub-divided housing became the market’s response to this new rising demand. It is interesting to note that the number of divorced persons in private rental housing has risen even more rapidly than that in public rental housing (see Figures 1 and 2).
Examine the Micro-Foundations of Socio-Economic Change
A low income divorced parent could apply for re-admission back to the public rental housing program often with preferential consideration if he or she had dependent children or remarried. These perverse incentives further tilted the balance in favor of divorce among low income families. It also exacerbated the demand for both public and private rental units. The additional housing demand from higher divorce rates was not well understood and was unforeseen by government and by society at large. Unfortunately this remains the case.
The high and rising divorce rate in Hong Kong is therefore both a cause and an effect of higher housing prices and rents. It distorts the measured inequality in household incomes leading to exaggerated alarms by the political left and social democrats about the rise of static measures of inequality. Their rise has origins in economic globalization and China’s opening. Rising property and rental prices fed on a rising divorce rate.
Unfortunately trickle-down economics took the blame for many false accusations of the failure of the capitalist economies. These economic and social changes were magnified through a public rental housing program that at best failed to protect the relative and absolute wealth position of families without property and at worst created perverse incentives that increased the divorce rate among the poor.
Rising divorce rates in the US and Europe have led to the decline of the middle-class as the poor descend into poverty with broken families that hold out less hope for their children. The effect of “no fault, unilateral” divorce laws and social welfare policies designed to help the poor has also led poor families to become more dependent on the state. The children of the poor in these affluent societies are losing their ability to become economically productive individuals capable of leading meaningful happy lives.
Our present day public housing program is reproducing poverty across generations against a background of sustained rising private property prices. I shall explore these effects in future essays.
Libertad Gonzáleza and Tarja K. Viitanenb, “The Effect of Divorce Laws on Divorce Rates in Europe”, European Economic Review, February 2009.
Douglas W. Allen, “Do No-Fault Divorce Laws Matter? A Survey, 1995–2006”, Unpublished Manuscript, Simon Fraser University, February, 2006.
Sixth essay in the series on Rekindling Hong Kong’s Magic