Wednesday, May 16, 2007

The Income Tax and the Single Female: The Case for Gender Based Taxation

A recent article in the Financial Times (April 18) written by two economists, Alberto Alesina and Andrea Ichino, advocated “gender based taxation”. The write-up is an extract from their research paper (March 2007). In the paper they make the case for differential taxation of income based on gender. The argument put forth is simple. Female labor supply is more “elastic” than that of males to after tax wage rates. Using “the Optimal Taxation Theory as the basic principle for allocating the tax burden “optimally”, they argued for “lower” tax rates on labor income of women, “higher” rates for men.

The proposal, irrespective of whether one agrees or disagrees with it, is timely since the US Congress (once our representatives get out of the Iraq war rut) is considering yet another change in the federal income tax code — a reduction in the minimum income tax and/or a change in the threshold. What makes the differential taxation issue worth revisiting is that it has surfaced THIRTY TWO YEARS after the famous Playboy Interview with then Secretary of the Treasury William Simon (May 1975).The Secretary, in one interview, was able to accomplish what many of us economists, tax experts and consultants have tried for many years — to reduce the tax rates on single person households. In the Playboy Interview, Simon used the phrase “the tax and the single man”. Some of you, the older generations, may remember that the Federal Income Tax had a tax rate on single persons that was 200% of that on married-couples households with the same income (the saying that “two people can live as cheap as one” did not sway the tax man). Differential rates much lower were imposed on other types of households — married filling separately, heads of households, and surviving spouse. Such tax treatment neither met any standards of equity nor was it based on efficiency measured by differences in the labor supply elasticity of single person or married couples. The Playboy Interview succeeded in prompting Congress to lower the rates on single persons, males and females. Without going through old issues, it suffices to say that since its inception the Federal Individual Income Tax has been evolving.

Enacted as an emergency measure during the Civil War, the tax was allowed to lapse in 1872 when the urgent need for revenues disappeared. For almost 20 years after the Civil War income tax, attempts at resurrecting the tax all but failed. The Sixteen Amendment to the Constitution ratified in 1913, gave a legitimate birth to the income tax. The Amendment gave Congress the “power to levy and collect taxes from whatever source derived”. The rate structure was progressive. Allocation of the tax bill among taxpayers under the guidance of the Sixteen Amendment would have meant that the FEDERAL INDIVIDUAL INCOME TAX SYSTEM does not DIFFERENTIATE AMONG EQUALS AND DOES SO AMONG NON-EQUALS. At the time, the tax unit was the INDIVIDUAL, HENCE THE NAME. The progressive rate was set in relation to income and income alone — not gender, not circumstances, or any other criterion. No good thing last for long! Over a span of 94 years and more than 20 reforms later, one would be hard pressed to find a semblance between the original intent and what the “reformers” have put on our plate.

With this brief history, let me return to the case being advocated for gender based taxation. Here we need to ask a couple of questions: is the proposal sound? Secondly, will such a proposal takes its “rightful” place in the tax code?
I will take up the second question first. My answer is unequivocally in the negative. It is unlikely to be acted upon at least in my lifetime. Reforms are slow and piecemeal. For over 25 years, I have written, testified before the US Congress House Way and Means Committee, consulted to the US Treasury Department about reforming the Federal Income Tax. I have learned through the years through such channels and experiences that reform is a slow process; it needs “powerful” advocates and persistence. And at the end what was advocated and what come to be are not one and the same. As the saying goes “you win some and loose some”. It is not only the PROCESS that makes implementation of reform lengthy but also the tax issues that have to be resolved are quite complex and far from easy. Think about the textbook principles of taxation which accompany any tax proposal: the tax and tax reform should be designed to meet three criteria — equity, efficiency and simplicity, a task beyond (thus far) the best efforts at reform (for a review see Ott and Vegari (2003), “Tax Reform Chasing the Elusive Dream”, in William Simon Contribution to Economic Policy, Atlantic Economic Journal, September 2003, 266-282). Even though the likelihood of passage of gender base taxation may be slim, it is worthwhile to see what it offers. The proposal is sound. The authors base it on two theses: taxes should depend on “non-modifiable characteristics of subjects that are related to their earning capacity” (Akerloff (1978), “The Economics of ‘Tagging’ as Applied to the Optimal Income Tax, Welfare Programs, and Manpower Planning”, American Economic Review, pp. 8-19), and taxes should be low on goods with high elasticity and high on goods with low elasticity (Ramsey (1927), “A Contribution to the Theory of Taxation”, Economic Journal, pp. 47-61; Mirrlees (1971), “An Exploration in the Theory of Optimum Income Taxation”, Review of Economic Studies, pp. 175-208).

What exactly “non-modifiable characteristics” means? The Federal Income Tax, levy taxes on US households. As mentioned above, the tax system classifies households for setting rates according to marital status and survivability. Clearly these are modifiable characteristics of the subjects. A single person household may become a married couple household; a married couple household may turn into a single person, a head of household, and so on. Thus, one needs to look for another characteristic — gender is a non-modifiable one. Gender based taxation fits Akerloff criterion. Note however that it does not state whether women should be taxed less, same or more than men. To differentiate we need the second theory.
Here I would like to pose to answer a question posted by someone to the Alesina/Ichino blog in relation to the optimal tax theory. The writer states: “As a complete and total non-economist, this is what I want to know: (1) is “goods with a more elastic supply should be taxed less” in fact a “general principle of public finance?” (2) What (briefly) is the force behind this general principle? How important is it, and why?”

These are very good questions. As a public finance economist, I shall briefly explain for the answer addresses the underpinning of gender based taxation. To (1), it is not a general principle of public finance, properly phrased it is a PRINCIPLE OF TAXATION, REFERED TO AS OPTIMAL TAXATION THEORY.
The theory was put forth back in 1927 by Frank Ramsey, who was asked by his professor the following question: If there are two goods, one has an elastic demand, the other inelastic demand and we want to raise a certain amount of tax revenues, what tax rates should you set on these goods, assuming that all persons have same preferences (utilities) for theses two goods (economists can’t answer questions without making simplifying assumptions). His answer was to impose a low tax rate on the good that is elastic and a high rate on the one with inelastic demand. Bravo. To see why this was a correct answer we need a bit more of economics. A good that has a high elasticity of demand (or supply) is characterized by two things: it has at least one substitute that is equally desirable and that a small rise in its price makes the consumer switch to the other good. If the switch is total, the good is said to have “infinitely” elastic demand or very high elasticity, if not total the elasticity is measured by the reduction in the quantity bought associated with the price increase. In short, an elastic demand for a good means that the reduction in the quantity bought as the price rise is quite large compared with the drop in the quantity associated with the same percentage change in the price of a good with inelastic demand. The same analysis applies to the supply. In the case of labor supply, the subject of discussion here, the price of offering one’s labor is the after tax wage rate. The change in the quantity of labor supplied due to a change in the after tax wage rate depends on a lot of things beside the wage rate (the measurement is a bit tricky), it depends on whether you can at least over the short hall get out of the market, cut the hours worked (under contract that may not be feasible), not enter the job market in the first place. Note again that a substitute activity has to exist. The substitute to labor is leisure, like fishing, playing golf, boating and so on. Leisure is valued by the after tax wage rate. Thus there are two steps. You either would enter the job market and the labor supplied changes, at least over the medium term with the change in the after tax wage rate or you do not enter the job market in preference for leisure.
Now we are ready to answer the question to whether the elasticity conclusions are correct, and why is it important to apply the optimal taxation theory — lower tax on the elastic supply relative to the inelastic supply. The response of the quantity to the price change based on the elasticity is correct. Since women labor supply is more elastic than the labor supply of men, a cut in their tax rate relative to men will increase their labor supply and income. Men, on the other hand have a highly inelastic supply so raising the tax on their wages will result in a very small drop in their labor supply. Why is this important?

Alesina/Ichino talk about “goals of society”, but for economists generally the reason given is “efficiency” of outcome. Economists use the term efficiency loss or excess burden to explain why the optimal taxation principle should be applied. The loss (which economists succeeded in measuring) arises due to the “tax induced behavior”. When the after tax wage falls and labor supply is reduced, the reduction is not voluntary — the individual is forced to substitute an activity that is not taxed to the one that is taxed. Since the choice is altered, there is a loss not only because work and hence output is reduced but also because of the substitution of a less desirable activity measured by the after tax wage rate for the activity (work) valued at the before tax wage. This is the efficiency loss. Think of having to use mass transport as the price of gasoline rises because of higher federal and excise taxes. In the case of labor supply, if the tax takes a big bite and your supply is elastic you might say the heck with it, I will go fishing. A working woman may decide to be a homemaker, a stay home mom, educate the kids at home and so on. The critical thing to remember here is that these were not the choices made before the tax is raised or imposed. Since labor is an important factor of production, when work is withdrawn capital or other factors are substituted and that (if you assume the best combination of factors were initially used) will result in less efficient outcome — output is lower and or prices higher. The loss to society is the sum of the loss in output and the loss associated with forced substitution of activities. If taxes are lowered on those with the elastic supply we get the opposite result. This is what optimal taxation is about (sorry economic theory takes a bit more space to explain even a simple idea).

The questions that need to be asked are: how does society value the input — the labor supply of women and labor supply of men? How each gender value its leisure time? And what valuation does society put on women’s alternative use of their time? I will not get into these issues. Alesina and Ichino (full paper) have a great deal to say about women making role models and children success and so on. They offer statistics to support their arguments for lowering the income tax rates on women (it is interesting that although single women have a lower elasticity of labor supply the proposal does not advocate an increase in their tax rates relative to married women!). To conclude, whether one accepts or rejects the case for gender taxation, and there is plenty on either side, it is well to remember that:
To strife for an optimal (efficient) tax structure in a many person economy, one needs to make specific assumptions about the nature of differences between individuals and the form of their utility or preference functions. Secondly, one needs to consider the implications of tax changes. As one reformer put it “when you close a loophole you end up opening a foxhole”.

Wednesday, May 9, 2007

To Be Eighty Years Young: Work, Health and Income of the Elderly

On April 27 2007, I was one of 200 invitees to a party to celebrate the 80th birthday and retirement of a member of the James Buchannan Center for Public Choice, at George Mason University. The feted lady, by all accounts and for those of us who are fortunate to know her is a remarkable person. She exudes warmth, efficiency and good southern hospitality. Whether at the society meetings or during my stay at the Center, I have always marveled at how such an efficient person could be so warm and caring for all of us visitors at the Center and not just to members of the Center and the University. On The 27th of April all of us who came to celebrate the day had no need to tell her how much her warmth have touched us all. On that day a major event had taken place for her and for us as well. She turned “Eighty”, and she has “Retired”, thus concluding her tenure at the Center. At eighty years young, she is as vibrant as she was at sixty, happy and in good health.

The celebration prompted me to revisit an issue that has been festering in my thoughts for some time: “AGING”. This issue has occupied, still does occupy economists, gerontologists, health care professionals, cosmetologists, policy makers and the public at large. The inevitability of aging is something that stares us in the face on a daily basis. Our society is obsessed about it. The health and beauty market is saturated with products that claim to restore our youth, products that let us “deify” aging and prolong our life. Yet when everything is said and done, no one “thus far” has escaped the aging process.

Over the past few weeks, Brian Williams the News Anchor of NBC have been exposing his viewers to snippets of the perils of aging in the “Trading Places” segment of the broadcast. The program opened with a view of how some aging family members of the network are dealing with aging. In putting before us the experience of their parents as how they cope with aging, the program succeeded in reminding its viewers that “we shall all be there”. Most importantly perhaps, the program brought to the forefront the dilemma facing adult children in caring for their aging parents. In another segment of the program the problem facing childless single parents was briefly highlighted. Given that sooner or later we shall all be aboard this train, it is perhaps useful to put forth some statistics before enumerating some of the problems facing the elderly.

It is a common practice to use the term “elderly” to refer to person 65 years old and over. As of the year 2000 in the US there were 32.6 million in that age group; 13.9 million males and 18.7 million females (this year is chosen for availability of data which will be cited here). Classified by marital status, single persons constitute 3.9% (4.2% male, 3.6% female); married with spouse present 54.6% (72.6% male, 41.3% female), widowed 34.2% (male 14.4%, female 45.3%).

What are the survival probabilities for this group?

One can assess survival probability for everyone aged or not from three types of information: age, race and sex. Expectation of life at an age in the bracket (67-70) is estimated to be in the range of 16.9-15.5 years; (71-75) in the range of 14.2-12.3 years; (75-80) in the range of 11.7-9.4 years; (80-85) in the range of 8.4-7.0 years ; (85-90) in the range 6.6-5.1 years; (90- 95) the range is 4.8-3.7 years. After the age of 95 it is somewhere between 3.5 and 2.5 years. The numbers are somewhat different when the population group is broken down by race, with the difference is more between white and black males than between white females and black females (data from National Vital Statistics Reports, vol54, # 14, April 19, 2006).
According to The National Vital Statistics, a male who was 65 years of age in 2003 would have ahead of him 16.8 years to enjoy; a female would have 19.7 years. Ten years later, at the age of 75 the years are 10.5 for males, 12.5 for females. At 80 years young, the numbers are 7.9 for males and 9.5 for females. The years dwindle down so that once the age 90 is passed it is but a few years left (unless you are an economist, as reported in THE ECONOMIST: “Economists live longer”).

Year in and year out, the data show an increase in life expectancy of both males and females with the largest increase recently recorded for black females. It is interesting to note that the much talked about “obesity” of the US population did not impact the survival probabilities. The 10 year probability of death between the years 1999-2002 has fallen for all age groups. Among those aged 55-74, it fell from 25.7% to 21.7% based on the findings of Cutler, Glaser and Rosen (NBER, program Report Winter 2006-2007). According to the authors, the risk profile of the US population is healthier now than it was bock in the 1970’s.

Longevity has good and not so good implications. The “Trading Places” stories shown on NBC, although small in magnitude to be statistically significant, nonetheless they give the impression that it is the latter than the former. It is good to be alive one hears, but few would couple the sentiment with at what price. The quality of life falls by the wayside, but when compared to no quality at all – what can one say; after all we have yet to discover the “quality” of dying. Perhaps someday we shall.

Having put down the life expectation statistics, there is a need to enumerate the problems associated with ageing. Clearly the first and foremost among these are the work and health status; the second is income or wealth, and the third is the life style and the delivery of care.
Let me first begin with few of my prejudices about labels. I do not much care for the labels “Elderly” and/or “Retired”. These labels project images of frailty and zero productivity. Let us think a bit about the work- income experience of persons 65 years and older. Retirement laws, written and unwritten employment codes influence if not often dictate retirement decisions of workers irrespective of health, ability to perform tasks assigned or productivity. At the work place, there is “a silent majority” voice for easing the “elderly’ out to enhance the upward mobility of the young and to economize on the wage bill. Given the statistic that a 67-70 years old who has already retired has a survival probability of 16-15 years how might one measures the value of life for such a person? And, given that we are healthier than three decades ago the chances are pretty good that such a person is likely to be as productive or nearly so as a person much younger in earlier decades. But society does not enter that into its calculus. For a non economist, the question may not be relevant even down right silly. But for economists this is as much as relevant as our valuation of risk to life (health and safety in the work place and or in cases of accidents). Value of life estimates (the courts use them to make awards) are made on the basis of the person age, education, income and work history. This value measures the contribution of the individual to society. If the individual is retired (assuming he/she does not have a second carrier) what measure should we use to calculate his/her contribution? I have always wondered why when a person is retiring it is assumed that he/she will be taking up fishing. This probably is an activity for which productivity can be calculated – the number of fish caught, the price of the fish, the value of the leisure associated with the act of fishing (assuming of course that one can make a catch). Of course there are non societal valuations of the retiree’ activities such as the contribution to friends and family as well as the contribution to oneself in term of enjoyment of leisure. But there is a diminishing utility to the consumption of leisure, when that is the only use of one’s time. Moreover, as with any non market activities valuations are individual specific. Since the “retiring” population in the years ahead is likely to increase in numbers, I would like to put before you the following questions: How should society valuate retirement as a non-market activity? Does it really matter for the individual and social welfare if a social value (positive or negative) were to be placed on a year of retirement? These are questions worth thinking about.

Being engaged in a non market activity does not mean that no income is received. Through the public and private pension systems and insurance plans, most retirees will receive pension income. There are numerous studies about retirement plans and retirement income and wealth. If the lack of income or insufficiency of income is not an issue facing a retiree, then the question of meeting need whether for living expenses or health care or nursing care would not arise. However there are segments of the populations 65 years and older whose needs currently are not being met and most likely will not be in the future. A quick reference is to look at poverty statistics.

In the year 2000, 9.7% of people 65 years of age and older were below the poverty level with the percentage much higher for females (11.8% compare to 9.7% for males). Add to this the fact that the majority of this age group are women without partners — single, divorced and widowed (56.1% of the total number of females), this places a further hardship on this group, not only to meet basic needs but also their future health care need. This is a subject that needs to be addressed collectively and individually. In reforming the health care system in the US, a subject that is likely to be taken up by the presidential candidates on their run to the White House, it is worthy of discussion to get informed about the future health care need of an aging population and the special needs group. Personal and social responsibilities for care have to be identified and choices of various options for the future critically examined. For the moment, two pieces of information are of note. In the year 2000, 4.5 % (1,557,800 persons) of the 65 years of age and older are in nursing homes. Of this total, 574,908 persons are in the age group 75-85, and 772,737 in the 85 years and over group. The cost of care varies according to type of accommodation, quality of care, location and severity of condition. A 2006 study by Sheila M. Loboprablu et.al. (Care Giver in Dementia: A guide for Health Care Professionals), put the cost of informal care at $257 billion, home care at $32 billion and nursing care at $92 billion. A Harvard Medical school study (2004), reports that the Massachusetts Division of Insurance put the yearly cost of nursing care at $76,000, while the average yearly Medicaid nursing care is put at $ 12,000 ( Care Giver’s Hand Book: A Guide to Caring for the ill, Elderly and Disabled—and Yourself). That is the good and bad of it. The good news, we live longer than our parents and grandparents; the not so good news is that without planning for longevity, getting quality care turn out to be an elusive dream.