Replacing Social Security with a Universal Basic Income

The idea of a Universal Basic Income (UBI) in place of existing entitlement programs (Social Security, Medicare, Medicaid, food stamps, unemployment insurance, etc.) that is financed by a flat rate income or consumption tax calls for a deeper discussion of its financing. “Our-social-safety-net”

For purposes of illustration, lets assume a UBI of 18,000 dollars per year for adults and half that for children (under 20 years old). This is somewhat above the current average Social Security benefit for someone retiring at age 65.  The current American population of 330 million consists of about 82 million children and 248 million adults. Thus, the total cost of such a UBI would be about 5.2 trillion dollars. This would exceed total expenditures in 2019 of 4.4 trillion dollars, of which 2.7 was for the social safety net (entitlements–social security and welfare). Total Federal tax revenue in 2019 was 3.5 trillion dollars of which 36% or 1.26 trillion dollars was from payroll taxes (social security and Medicaid and Medicare). This left a staggering deficit in 2019 of almost one trillion dollars that had to be borrowed from China and others when our economy was at full employment and should have been in surplus. And now look at our shocking deficit in this pandemic year! But that is another story.

The goal of this note is to illustrate the significant progressivity that would exist with a flat rate consumption tax in place of the corporate and personal income taxes and the payroll tax when replacing existing safety net expenditures with a UBI.

The original rational for the regressive payroll tax to pay for social security pensions was that social security was a traditional retirement program funded from the savings of each pensioner. The assumption was that the pension it paid reflected the money that each worker paid into the so called but misnamed social security trust fund. In short it was characterized as what we call a defined contribution system (you get what you saved) when it was in fact structured as a defined benefit system (you get a defined amount no matter what you actually paid in). In fact, as Americans lived longer and longer, thus enjoying more and more retirement years, the system collapsed into a basically pay as you go system (today’s workers were paying for today’s retirees’ pensions). The trust fund has very little savings in it. So instead of the payroll tax funding the worker’s future pension it became a regressive tax funding current retiree pensions. The payroll tax should be abolished. “Saving Social Security”  

A flat tax (whether on income or consumption) has many economic virtues, with simplicity at the top of the list.  But a flat rate rubs some people the wrong way who think that the wealthier should pay more tax than implied by a flat rate. My sense of fairness calls for someone with twice the income (if we are focusing on an income tax) to pay twice the tax. That is exactly what a flat rate tax (the same tax rate for everyone whatever their income level is) provides.  Whether we go for an income tax or a consumption tax we should forget about the corporate income tax. It is more trouble than it is worth in a world in which many if not most companies operate globally (i.e. in many different tax jurisdictions). It only contributed 7% of total Federal tax revenue in 2019 and unfairly taxes the company incomes of company owners twice. “Principles of Tax Reform

A flat personal consumption tax that would raise the same revenue as raised by all Federal taxes in the U.S. in 2019 (3.5 trillion dollars) would require a 24% rate. But that revenue did not cover all of the government’s expenditure as noted above. In addition, replacing existing safety net expenditures of 2.7 trillion dollars with a UBI of 18,000 dollars per adult and 9,000 dollars per child (5.2 trillion dollars) would result in total Federal government expenditures in 2019 of 6.9 trillion dollars or 2.5 trillion dollars higher (5.2 – 2.7). A flat consumption tax rate of 47% would be needed to raise 6.9 trillion dollars. This seems very high for two important reasons. First it assumes a balanced budget for the actual level of defense and other non-entitlement expenditures in 2019, i.e. it raises almost one trillion dollars more than the government actually collect in 2019.  Secondly it is financing the additional 2.5 trillion dollars for the UBI from which the higher tax would be paid, i.e. the net tax would be lower.

Though the marginal tax rate would be flat (the same for everyone), the resulting tax burden would actually be quite progressive. To provide a rough idea of the net progressivity of the UBI with a 47% consumption tax, assume that all income is consumed (this would overstate consumption some for higher income families). The poorest families, those who have no income other than the UBI, and assuming a family of two adults and two children, would pay no net tax and receive a net income subsidy of $28,620 or $2,385 per month. For the median family in 2019 (50% income level) the average income was $63,030. But including their UBI their total tax payments would be $55,004 or an excess of $1,004 over their UBI for a tax rate on their (pre UBI) income of 1.6%. For the family at the 80% income level, the average income was $130,000 and their total tax payments would be $86,480 or an excess over their UBI of $32,480 for a tax rate on their earned income of 25%.  This is a significantly progressive outcome while preserving the flat marginal rate.

Individual states may well choose to provide assistance for individual specific purposes, as they do now, for example, for education at various levels. But at the Federal level every effort should be made to prevent such add-ons to our social safety net. If as we become richer as a whole, we choose to be more generous, the amount of the UBI can be raised (or lowered) but only for everyone equally. This would prevent individual interest groups from tacking on special assistance for themselves. The ability of such special interests to gain special favor is a major reason for the slippery slope of the creeping welfare state we now enjoy.

It is important that policies, whatever their good intentions might be, also provide good incentives for outcomes that are desirable for society as a whole. A danger with progressive marginal tax rates, is that the majority of taxpayers have an incentive to raise the rate on those with incomes greater than their own. Or at best there is no incentive for them to resist such a temptation. Soaking the rich is not only an unfair treatment of those who have prospered inventing and delivery goods and services we liked enough to buy, but it will drive them away to tax jurisdictions that better respect their property rights. One of the many virtues of a UBI financed with a flat rate consumption (or income) tax is that the only way to increase the average tax rate on the wealthy is to increase the UBI for everyone.

A UBI would fulfill our desire to help those in real need but would return the responsibility of individual decisions of how that assistance is to be used to the individuals involved. It would simplify and depoliticize the determination of who gets help and how much and would remove the burden of determining our proper tax obligation for most of us. It would thus greatly simplify the administration of such a combined program. It would better align the political incentives for the level of assistance with the preferences of society as a whole. While most people work for more than the income it generates–the self-esteem of being a useful member of society is also important–a UBI would remove the economic disincentive of many current welfare programs of working resulting from the loss of benefits when income reaches a modest level.

Our Social Safety Net

Virtually every country provides assistance to their poorest citizens–to those who fall for one reason or another from normal employment. Approaches to fulfill this objective differ widely. The Federal safety net spending in in the United States in 2019 was 2,600 billion dollars. Total government (Federal, State, and local) entitlement spending was 2,900 billion dollars. Of this, about one third was for Social Security and one third for Medicare and Medicaid.

“The federal government funds 126 separate programs targeted towards low‐​income people, 72 of which provide either cash or in‐​kind benefits to individuals. (The rest fund community‐​wide programs for low‐​income neighborhoods, with no direct benefits to individuals.) State and local governments operate more welfare programs.”[1] This year, in response to the Covid-19 pandemic, the Federal Government has added about $3 trillion dollars ($3,000,000,000,000) for temporary one-time assistance for the impact of the forced interruption of production, and is likely to add more.

The goal of these programs is, or should be, to adequately support those needing it without creating disincentives to work and with minimal abuse (corruption). The CARES Act and other pandemic assistance programs were quickly created in an emergency. It is thus understandable that mistakes were made. As time goes on charges of corruption (politically motivated expenditures) are multiplying. The administrative challenges of suddenly making millions of individual payments quickly and correctly were and are huge.

There is a dramatically better way to do this. We might characterize our existing approach of government directed assistance (e.g. food stamps) as the Socialist Model. It is top down and dictates how the assistance is to be used. Replacing all of these programs with a Universal Basic Income (UBI) leaves the decisions on how its recipients use it with each individual. This might be characterized as the Individualist Model. It has many advantages over our existing approach.

A UBI would eliminate all government discretion over who receives assistance and how much they receive.  Every adult citizen would receive that same amount monthly (and every legal child would receive the same smaller amount). The government could not favor one group over another on any bases other than age. This removes political considerations from defining and administering the payments. Every birth and death in the country is recorded in a county hall of records and every legal immigration is recorded with the United States Citizenship and Immigration Service (USCIS). Thus, the records upon which payments would be based already exist. Using them would remove my need every year to submit a certified document to my pension plan stating that I am still alive. No special measures or supplements would have been needed to address the personal income shocks of covid-19.

One carveout would be required for medical insurance. Every person or family should be required to use part of its monthly UBI to buy a health insurance policy that at a minimum includes catastrophic care coverage. At the launch of UBI and this insurance mandate, insurers would not be allowed to refuse coverage to people with preexisting conditions and the government would cover the actuarial estimate of the extra cost of such conditions.

As the UBI would replace Social Security pensions, another modification would be for the unlikely case that the UBI would be less than ones current Social Security pension (which in 2017 was $13,824 for someone who retired at age 65). So, for such a person who has already retired or is, say, within five years of retirement, their UBI should not be less than their existing Social Security pension. But a UBI of $1,500 per month ($18,000 per annum) for everyone seems reasonable, which would make this case mute.

Those with a welfare state mentality argue that people can’t be trusted to spend such income wisely (from their perspective). I reject such thinking. There are among us, of course, those individuals with addictions and mental illnesses who are indeed not capable of making their own decisions and thus caring for themselves. Our laws and practices already provide for such special cases and would continue to supersede the rights of such individuals to make their own decisions about the uses of their UBIs.

But can we afford it? Today’s American population is about 330 million, of which about 80 million are under the age of 20. To get a rough sense of what is possible, if we replaced today’s safety net expenditures of 2.9 trillion with a UBI to the 33% poorest (110 million people, of whom 25% or 26 million are under 20 years old) in the U.S. in 2019 each person could received about $26,000 per year or about $2,200 per month. If children (those under 20) are paid half what is paid to adults, existing safety net expenditures could finance about $15,000 per child per year and $30,000 per adult if a UBI is given to the lowest third of the population in terms of income. For a family of four this would be an annual income of $90,000, which is above the median household income of about $64,000 in 2019, and is clearly excessive.

But a UBI must be universal. It must be paid to everyone for several simple reasons. Most importantly, it would eliminate the disincentive to work in the existing programs, which end if a person’s income rises above a specified level. With a UBI, every additional dollar of income joins an irreducible UBI. Every additional dollar earned make the recipient that much better off (the UBI amount plus the earned income). This is a very important feature. In addition, it would remove any political question over the level of income at which it should be withdrawn. It would be paid to everyone regardless of their income.

But paying the UBI to everyone, not just the lowest one third, would triple its cost. Can we afford it? Clearly those who pay taxes will have to pay more to cover this additional cost. But as the additional cost is to cover payments to these taxpayers, they would not pay more on net (depending on the nature and structure of our tax systems).  For reasons of equity and tax efficiency I have long advocated a flat tax, meaning the same marginal tax rate for everyone paying taxes. “My-political-platform-for-the-nation-2017” I would go even further and replace income taxation (both corporate and personal) with a flat rate, comprehensive consumption tax.  When a flat rate tax, whether income or consumption, is combined with a UBI, the net result is mildly progressive. Low income people pay no tax on net and in fact receive net income via the UBI (what Milton Friedman called a negative income tax). Middle income families might break even (receive a UBI sufficient to pay for their extra taxes) and for higher income families their extra taxes would be greater than their UBI, hence a progressive average tax rate system even with flat marginal rates.

A UBI with a flat rate consumption tax would enormously simplify our tax and welfare system while improving the financial incentives to work and returning more control over our lives to individuals from the state. Covid-19 dramatically demonstrates that the time has come to replace existing welfare systems with a Universal Basic Income (UBI).

[1] Michael Tanner, “When Welfare Pays Better than Work” CATO Institute, August 19, 2013.

Tax reform and the press

I have written several articles about the need for serious tax reform in the U.S. and set out the basic principles of good tax law accepted by most economists. “US Federal Tax Policy”, Cayman Financial Review, Issue 16, Third Quarter 2009.  Cayman Financial Review, July 2013

Taxing consumption is best, but if income is taxed, it should be broadly defined and taxed uniformly. Income tax reform should follow the mantra “broaden the base and lower the rate” for the revenue needed to finance whatever the government spends. The main dispute tends to focus on whether and how progressive the tax rate should be. I favor a flat rate as the fairest and simplest regime. This means that a person with twice the taxable income would pay twice the tax. Many others favor a progressive rate—a marginal tax rate that increases with income—, which means that someone with twice the taxable income might pay 3 or 4 times as much in taxes. In 2016 the “top 1%” by income paid over 50% of all federal income tax revenue collected and the top 20% paid 84%.

I raise this issue because any judgment of whether a reduction of the top U.S. marginal tax rate from its current 39.6% to 38.5%, as currently proposed by the U.S. Senate, increases or decreases the fairness of the system depends on whether you consider 39.6% fair or too high or too low. I consider it too high and a reduction to 38.5% too little, so I would say that the tax reform is unfair to the top income groups by not lowering the top tax rate enough. The press almost uniformly refers to any cut in the top rate as favoring the rich (rather than reducing discrimination against the rich).

But what prompted this note was the blatant bias reflected in the following Washington Post article that claims to report the winners and losers in the current Senate tax reform proposals. Winners-and-losers-in-the-Senate-GOP-tax-plan   In the losers column the article states the following for the poor:

“The poor. More than 70 million Americans don’t make enough money to have to pay federal income taxes. Many of those people currently receive money back from the government because they qualify for refundable credits. Under the Senate plan, those credits aren’t going away, but they also aren’t growing. On top of that, the plan raises America’s debt, which will likely require cost cuts somewhere down the line. Republicans have proposed sizable cuts in the past to some safety net programs used by the poor.”

According to the author of the article, Heather Long, the poor lose because they don’t gain anything!!! Seventy million of them don’t pay taxes to begin with so there is not much that tax reform can do to lower their taxes. The existing tax credit paid to these people will remain but is not increased. Thus Heather concludes that the poor are losers because they didn’t gain anything. I agree with Heather’s implicit objection to the plan’s increasing the federal government’s debt, but avoiding that would require higher taxes for someone and has nothing to do with making the poor worse off that I can see.

Any tax reform that is revenue neutral (unfortunately this one will increase the debt by 1.5 trillion dollars over ten years.) necessarily increases taxes for some while lowering them for others.  It should not be judged by whether it will result in President Trump paying more taxes or less, as some press would have it.  It should be judged by whether the resulting realignment of tax obligations is fairer and economically more efficient (neutral). Sadly it is rarely discussed in these terms.