(This essay was published in Hong Kong Economic Journal on 30 December 2015.)
The Poverty Commission has just released its long-awaited policy consultation document Retirement Protection Forging Ahead. The public consultation process was plunged into controversy before it began by leaks that the government decided to frame the retirement protection options as “regardless of rich or poor” or for “those with financial needs”. The powerful and vocal lobbies for “universal retirement protection” are incensed. This is not surprising. It is an issue that deeply divides society.
The government proposal contains some interesting simulations that compare the budgetary outcomes of granting a monthly payment of $3,230 (in 2015 dollars) either to all elderly individuals (defined as above 65 years of age) under the “regardless of rich or poor” option or to “those with financial needs” (defined as having maximum assets of $80,000 for a single person and $125,000 for a couple). A comparison of these two proposals with the existing arrangements (which is called the “baseline scenario”) is presented in Table 1.
The elderly social security expenditure under the “baseline scenario” increases from $25.3 billion in 2015 to $49.8 billion in 2064 as the elderly population grows (equivalent to a growth of 1.38% per annum). It increases from $27.9 billion to $55.8 billion (equivalent to a growth of 1.41% per annum) under the “those with financial needs” option and from $47.9 billion to $106.1 billion (equivalent to a growth of 1.62% per annum) under the “regardless of rich or poor” option. The respective annual growth rates can be interpreted as being driven by the population ageing.
These three growth rates are not very different. But if elderly social security expenditures start from a higher base level in 2015, then they will also end up at a higher level in 2064. For example, in the year 2064 the “those with financial need” option would end up spending 121% more than the “baseline scenario” in 2015, while the “regardless of rich and poor” option would be spending 319% more. The former implies an annual growth rate of 1.61% and the latter 2.92%.
It appears to me that the government’s estimated figures throughout the period 2015-64 assume a constant level of real income per population. This implies that the government calculations are sensible and can be meaningfully interpreted if monthly payments were to be adjusted proportionally with increases in real income per population over time.
The simulated calculations for raising tax revenues to fund the expenditures are deeply concerning regarding the long-term viability of the “regardless of rich or poor” option: either (1) an increase in profits tax from 2.8% in 2015 rising to 4.6% in 2064, or (2) an increase in salaries tax from 4.9% in 2015 to 9.1% in 2064, or (3) introducing a goods and services tax of from 2.8% in 2015 to 5.0% in 2064.
The issue at stake is not affordability. One can always choose to spend less on oneself and transfer more income to social security of the elderly. But one has to worry that an additional 319% increase on social security expenditure alone would impose a great deal of stress on our economy if we wish to preserve the robustness of our fiscal system, our linked exchange rate regime, and our competitiveness.
There will be even more pressing demands to spend on health care, which will grow at an even faster rate than elderly social security expenditures in the coming decades. These too will require higher tax rates unless we choose to provide a lesser quality of health care to the elderly.
One has to recognize that, on the one hand, there are those who are committed to the preservation of a limited government, low tax rate regime, and free open economy as the foundation of our economic successes and civic freedoms. They view the creation of a universal retirement protection scheme as a grave threat to our freedoms and prosperity. If the taxpaying population is called upon to pay for the general elderly public, one has to consider what this would do to work incentives and savings behavior.
One should not ignore the fact that if society begins to subsidize everyone in old age, it will alter everyone’s propensity to save for old age. Consider the effect of rising government social welfare expenditure on work incentives. Figure 1 plots the percentage of the population not in the labor force for no compelling reason against the inflation-adjusted government social welfare expenditure per capita. In every age group from age 20-59, the percentage of population choosing not to work increased from less than 1% to over 3.5% in a span of two decades.
Hong Kong has a severe labor shortage problem and yet we are subsidizing people to stay out of the workforce. If our society continues to destroy work and savings incentives, our economy will not be able to prosper, probably with very detrimental consequences for our competitiveness.
Moreover, our commitment to a linked exchange rate regime imposes severe discipline on fiscal spending to maintain the credibility of our monetary and financial system. This is a point stressed repeatedly by Harvard economist Dani Rodrik, who demonstrated that the requirements of a fixed exchange rate regime (like the system in Hong Kong) must not accommodate domestic social demands for that would produce a fundamentally incompatible contradiction.
If public expenditures rise continuously and public debt levels climb to meet populist social demands, then there would be growing inflationary pressure to devalue our currency. At that point it would not be possible to maintain our commitment to the fixed exchange rate. Professor Rodrik calls it the inescapable trilemma of the world economy, which he calls the “impossibility theorem.”
Introducing a universal retirement protection scheme now, especially when Hong Kong has a rapidly ageing population and rapidly rising pressure to spend on healthcare, could unravel our exchange rate regime if government were to give in to populist political pressure.
On the other hand there are those who are committed to serving their constituencies among the elderly. The past three decades have been a period of rapid economic transformation that has created a more unequal society. Growing segments of our population now live in poverty amidst prosperity.
As a result of government policy under Donald Tsang, many elderly households were able to move into public rental housing, which today accommodates more than half of elderly households. Meanwhile, the elderly poor living in private housing are highly concentrated in a few neighborhoods. The concentration of elderly and poor households has made the old age poverty cause politically popular and their constituents easier to mobilize.
At the same time many in the middle income classes, especially those in the lower-middle income groups, are finding it more difficult to achieve a comfortable standard of living. The rising cost and share of housing, health care, education, and life expectancy in a household’s lifetime budget expenditure has made them feel nearly impoverished. Universal retirement protection holds out the promise of alleviating their contributions to the upkeep of their ageing parents.
Since only half the working population pays taxes at present, this is indirectly a transfer for the benefit of many lower-middle income working age households who pay very little tax and will not be burdened heavily even if tax rates were to rise.
Although humanistic relief for the poor and compassion for the elderly are ethical principles embraced by our community, the two ideas are not the same. Unfortunately when they are amalgamated into a single concept it can produce a compelling and powerful political force for supporting universal retirement protection. It is important to unbundle these two ideas if we are to have a coherent policy.
Our elderly poor need relief and the “those with financial need” option makes the best sense. Our other elderly population needs support in health care. I have always believed the less well-off children of our elderly are best served if we can help their parents living in government-subsidized housing to become owners of their rented homes, which their children could some day inherit. Fostering a more robust intergenerational contractual arrangement between relatives is much better than one among strangers. If compassion cannot be fostered within the family, then it would be a stretch to believe it would thrive in the community at large.