Center for Business and Public Policy
at the University of Illinois, Urbana-Champaign

Welcome to my first posting on the CBPP blog.  I am honored at the invitation to join the list of distinguish contributors and hope to continue the tradition of providing informative and thoughtful content to our readers.  Now on to my post!

The editorial in Tuesday’s Daily Illini titled “Sustainability investment guarantees return” extols the virtues of the newly increased student Sustainability Fee.  Back in 2008, the annual $5 Fee was created by student referendum to fund projects that “help establish a sustainable campus environment by financing initiatives such as green buildings, engagement of the university community, recycling, energy efficiency, and environmentally responsible purchasing.”  This past year another referendum raised the Fee to $14 per student per year.

The Daily Illini’s editorial claims that Fee helped pay for a $450,000 lighting efficiency project at the Krannert Center that would save $70,000 annually for the next 20 years.  Assuming that these figures are correct, the implied 14.5% Internal Rate of Return (IRR), while not overwhelming, is a solid return on a capital investment particularly given current macroeconomic circumstances.

The existence of the Sustainability Fee raises two questions that I want to address.

(1) Why the need for a student Sustainability Fee if the projects it funds provide such good financial returns to the University of Illinois?

For readers of this blog and for general members of the University community, the answer is probably self-evident: budget problems.  The State of Illinois and the University can barely (and sometimes not) cover current operating expenses, leaving no room for projects with large upfront costs that provide future benefits.  That is, there is no money for long-term investments.  I am not saying that all projects funded by the Sustainability Fee will provide at least a 14.5% IRR, but one can envision many other energy efficiency projects on campus that could yield high returns that go wanting for lack seed money.

(2) Why would the current generation of students want to impose costs on themselves when the majority of benefits accrue to future generations of students?

Looking at the 2010 referendum results, the $9 increase in the Sustainability Fee passed with 77.1% of the vote.  While that seems like an impressive margin, only 13% of the student body actually voted on the referendum.  This means that the 3,885 students who voted “Yes” for the Fee increase imposed over $300,000 in costs per year on “No”-vote  and non-voting students.  (To be fair, the Sustainability Fee is refundable upon request; however, the default opt-in and in-person refund process probably leads to high Fee participation.)  Thus, my first point is that not all of the students choose to impose the Fee.

Next, since undergraduate students are only on campus four years, any project with more than a four year payback period will not be financial beneficial to the students.  That is, the generation of students that paid for the Krannert lighting project receives a NEGATIVE rate of return ($450,000 invested with 4 X $70,000 in net savings).

So what is going on here?  It seems true that university students are more concerned with the environment than are the general population, and probably more so for the subset of students that self-selected to participate in the referendum.  Therefore, the students voting “Yes” for the Fee could be gaining non-monetary benefits from the sustainability projects (such as the “warm-glow” of doing the environmentally responsible thing).  However, I have another, potentially more nefarious explanation: their parents’ credit card.  To the extent that parents pay for room, board, and fees for their children, the ability for students to impose the Fee for a cause they feel “good” about is just another way of spending their parents’ money!

I hope you all have a fun and safe Labor Day weekend, see you all next time.

State lotteries collect more than they pay out, in order to raise net revenue for the state – an implicit tax.  For example, you might pay $1 for a chance to win $100.  But if 200 people buy such tickets, then the state gets to keep 50% of the take.  The chance to win is only a half of one percent, so the expected value of the ticket is only 50₵.

Much of public economics is concerned with who bears the burden of such taxes.  Any tax that takes the same fraction of income from everybody is said to be proportional, whereas the income tax that takes higher fractions of high incomes is said to be progressive.  In contrast, a tax on electricity would be regressive, because electricity constitutes a higher fraction of low-income budgets than of high-income budgets.

What about the implicit tax on lottery tickets?  Just like electricity, lottery tickets constitute a higher fraction of spending for low-income households than for high income households.  The book about state lotteries by Charles Clotfelter and Philip Cook has a very appropriate title, “Selling Hope” (Harvard University Press, 1989).   As a result, the implicit tax is said to be regressive – a burden that’s a higher fraction of income for the poor than for the rich.

 I’ll claim here, however, that state lotteries are progressive, not regressive.  They provide a benefit to low income families that is a higher fraction of income than for high income families.  The key to this point is to think about lotteries “relative to what?”.  The usual question about a tax considers the burden on high and low income families relative to no tax. Indeed, if the state allowed a competitive market for lotteries with no tax, then the average payout would pretty much match the sales revenue minus normal administrative costs, with no net profits.   Relative to that alternative, the lottery tax is regressive (takes proportionately more from the poor).

But a world with many untaxed competitive lotteries is not the relevant alternative!  The question over the past few decades for each state legislature is whether to allow lotteries at all!   Relative to that alternative, no lotteries, the introduction of a lottery – even a taxed lottery – provides people an opportunity to buy something worth more to them than its price.  This difference is what we call consumer surplus.   If you are very hungry, you might be willing to pay twenty dollars for a hamburger, if you really had to do so, but fortunately for you, the price is only $5!

The introduction of a product provides consumer surplus to those who buy it, since it is worth at least the price to them and probably more than the price they have to pay.  For lottery tickets, those who spend a higher fraction of income on lottery tickets are the poor, and so the introduction of lotteries provides benefits to the poor that are a higher fraction of their income than for others. 

You may object to that logic, but let’s figure out what might really be wrong with it!  That logic is absolutely unassailable to anyone who buys into consumer sovereignty and the economic efficiency of the free market.  That logic says consumers are fully informed and will buy whatever makes them happy, with no negative effects on anybody else.  If so, then the poor benefit from lotteries.

Any valid objections to my logic above would have to explain exactly what is the market failure.  Do buyers not know that the state is keeping some of the money, and therefore the expected payout is less than the price of the lottery ticket?  I think they do understand that.  Are they imposing negative externalities on others?  Single individuals should be allowed the freedom and responsibility to decide for themselves how to spend their money.   A married couple with kids might want to plan a budget together, but there’s no inherent reason that budget couldn’t include a little fun with lotteries.  Some behaviors might be pathological, like alcohol or gambling addictions, but that is not a reason to ban all alcohol or gambling.  It is a reason to watch out for the pathological cases, and try to help out those individuals.

I certainly don’t mean to claim that lotteries have no issues at all.  Public policy might want to deal with those issues.  But most people who want to ban lotteries are being paternalistic, with the inherent belief that they know what’s best for other people.  They want to impose their beliefs on other people.  That’s not consistent with a free society.  If people are allowed to choose for themselves, then lotteries make poor people better off.

I saw an interesting piece in the New York Times entitled “Debts Rise, and Go Unpaid, as Bust Erodes Home Equity.”  Home equity loans are secured against homeowners’ equity in their homes.   During the housing boom, it was not uncommon for a homeowner to buy a home for, say, $100,000, watch its “value” increase to $200,000, and then take out a loan secured against this increase in the value of the home.  Just by riding housing values up, this homeowner could gain access to a line of credit of up to $100,000.

If housing values continue to rise and people keep working, everybody is happy.  Homeowners can make relatively small payments on the home equity loan and enjoy increased consumption in the short run.  Eventually, when they sell their house at a higher price, they can pay off the loan and everyone wins. 

However, in the face of a national recession and housing bust like we’ve experienced lately, things look quite different.    People have lost their jobs, and so can no longer afford to make their loan payments.   In ordinary times, the lender would seize the collateral for the loan – in this case the house.  However, at the same time we’ve been going through a recession, we’ve also been experiencing a housing bust.  So, the house that had been valued at $200,000 at the time the loan was written may not only be worth $80,000.  The collateral is no longer there.

The result of this dynamic is an increase in defaults on home equity loans.  Faced with financial difficulty, buyers are choosing not to pay their home equity loans and challenging the banks to try and collect.  However, in the case of home equity loans, this can be particularly difficult for banks, since, following bankruptcy, home equity loans are paid off only after primary mortgages.  So, if an equity lender tries to collect, the borrower can simply threaten bankruptcy, in which case the equity lender will most likely get nothing.  Thus, rather than being at the mercy of the banks, households that took out large amounts of debt are actually in a position of power.  To quote a couple of paragraphs from the article:

The result is one of the paradoxes of the recession: the more money you borrowed, the less likely you will have to pay up.

“When houses were doubling in value, mom and pop making $80,000 a year were taking out $300,000 home equity loans for new cars and boats,” said Christopher A. Combs, a real estate lawyer here, where the problem is especially pronounced. “Their chances are pretty good of walking away and not having the bank collect.”

Even when a lender forces a borrower to settle through legal action, it can rarely extract more than 10 cents on the dollar. “People got 90 cents for free,” Mr. Combs said. “It rewards immorality, to some extent.”

This phenomenon points toward an interesting change in American culture that reminds me of a story from my own life.  When I was in graduate school, I would rent movies from Blockbuster Video. (This was pre-Netflix, but at least they were DVD’s!)  I used to take the due dates very seriously, to the point where I’d pull on my shoes and run out at 11pm just to return a movie before the midnight deadline.  Then, one day, I didn’t, and I realized that nothing bad happened to me if I did.  Sure, there was a fine, but it was one I could deal with.  From that day on, I don’t think I ever returned another video on time.

What we’ve been experiencing lately with regard to bankruptcy and loan delinquency is much the same.  There hasn’t been a change in bankruptcy policy.  And, while there has been a change in economic circumstances (some responsible people can’t afford to repay reasonable loans that were taken out in better times), there has also been a cultural change whereby declaring bankruptcy is no longer seen as a last resort.  It has become acceptable to declare bankruptcy strategically, even before all available options for repaying as much of a loan as possible have been exhausted.  And, lenders are now aware of this.

It is difficult to know the impact of this change on lending markets, but it is likely to be profound.  Home equity loans, which used to be straightforward, are likely to be more difficult to acquire, as lenders begin to protect themselves against the possibility of a housing downturn in a world where people feel it is acceptable to walk away from their obligations.  These protections will likely involve fewer loans, smaller loans, higher interest rates and stronger collateral requirements.

This coming Saturday, August 14, marks the 75^th birthday of the U.S. Social Security system. Specifically, it marks the date that President Roosevelt signed the Act into law, famously stating:

“We can never insure one hundred percent of the population against one hundred percent of the hazards and vicissitudes of life, but we have tried to frame a law which will give some measure of protection to the average citizen and to his family …”

The original Act specified that benefits were to be paid only to primary workers when they retired at age 65.  The Act established that benefits would be based on payroll tax contributions made during the working years.  Of course, the program has been modified many times over the years (e.g., allowing benefits to be taken at 62, expanding coverage to spouses, disabled workers, and others, dramatic increases in tax rates, changes in benefits, etc). 

Initially, benefits were paid as a lump-sum.  While Ida May Fuller is best known as the first recipient of Social Security benefits, SSA’s historian indicates that the first benefits were paid as a lump-sum, and that:

“The earliest reported applicant for a lump-sum benefit was a retired Cleveland motorman named Ernest Ackerman, who retired one day after the Social Security program began. During his one day of participation in the program, a nickel was withheld from Mr. Ackerman’s pay for Social Security, and, upon retiring, he received a lump-sum payment of 17 cents.”

It was not uncommon for early recipients to receive much more than they put in.  Indeed, it has been estimated that the net transfers to early generations of recipients is well in excess of $10 trillion.  In other words, for most of the last 75 years, the majority of Social Security recipients received far more in payments than they paid into the system (and, yes, this is true even if one accounts for inflation and implied interest on those contributions.)

How is this possible?  Actually, it is quite simple.  Social Security is not a funded pension system.  It is a “pay-as-you-go” transfer system in which the funds paid out to current beneficiaries are provided by current taxpayers.  Such a system can work quite well so long as we have wage growth and so long as the ratio of workers-to-retirees is stable or growing. 

But therein lies the crux of Social Security’s financing problems.  Unlike what many citizens believe, the true problem facing Social Security has very little to do with Congress’ penchant for “spending the Social Security surpluses” of the past 25 years.  It has far more to do with the basic financing structure of the program.

In the 1950s, there were 16 workers paying taxes to support each Social Security beneficiary.  By the time JFK was elected President, it was about 5 workers per beneficiary.  Today we have a bit more than 3 workers for each beneficiary.  In my lifetime, that will fall to 2 workers per beneficiary.

So do the math.  If you want to replace 40% of the average workers income upon retirement, and you have 16 workers supporting each retiree, you only need to collect taxes from each worker equal to 2.5% of their income (2.5 x 16 = 40).  With only 5 workers per retiree, you need to tax them at a rate of 8%.  When there are only 3.3 workers (today’s ratio), you need a tax rate of 12.1%.  (Today’s combined tax rate is about 12.4%).  As the ratio falls to 2-to-1, tax rates need to climb to 20% to keep the system in balance.

(I am simplifying a bit here, but it is remarkable how closely this very simple calculation mirrors the Social Security Trustees’ long-term financial outlook!)

So, as we celebrate the birthday of the Social Security system, we have to ask ourselves some difficult questions.  Can we afford the system we have?  If not, whose benefits do we cut? High income retirees ?  Low income retirees?  Today’s retirees?  Today’s workers?  Alternatively, whose taxes do we raise?  Everyone?  Only high income households?

Just as most members of the human race who are fortunate enough to live to age 75 begin to notice varying degrees of health declines due to aging, so too must we deal with the unhealthy economic consequences of an aging Social Security system. 

The Social Security Disability Insurance (SSDI) program is an important part of the social safety net in the U.S.  If ever there were a risk that ought to be insured, it is the possibility of experiencing a physical or mental disability that brings one’s working-life to an end.  Those of us that have loved ones who rely on the SSDI program as a major source of household income understand how important it can be to financially sustaining those who are unable to continue working.

But the program can also be criticized in many ways.  First, the backlog of cases is very high – meaning that those who are disabled must often a very long time – sometimes even years – before they receive their first check.  There has also been a tremendous rise in the SSDI program caseload, which is placing enormous financial strain on the program as well as on the Social Security Administration’s already stretched field offices.

Nearly all of these problems trace to one root cause – that there is no simple test for determining who is truly disabled, and who is just trying to pass themselves off as disabled so that they can receive monthly checks for the rest of their lives without working. 

I know, some of you are going to say, “who would possibly do that?”  Indeed, some are offended by the notion that any undeserving individual would attempt to “act” disabled when they are not. 

But let’s be honest.  If it were easy to determine who was disabled and who was not – if there were some simple and fool-proof blood test or lie-detector test – then there would be no need for a huge bureaucracy of SSA claims reps, no need for 50+ state disability determination units, no complex layers of case reconsiderations and appeals, no need for hundreds of Administrative Law Judges, and no delays in processing checks.  There would be no backlog of cases.  And, frankly, there would probably be a lot more willingness among the general public and elected officials to generously support the program. 

But it is not that easy.  When a person argues that their back pain or mental condition means that they will no longer be able to work, the law requires that Social Security determine whether the person is indeed unable to earn more than the “Substantial Gainful Activity” amount each month – not just in their prior job, but in any job.  They must also determine whether the disability is permanent and/or likely to result in their death.  No easy task.

Ultimately, however, it is an empirical question whether there are people who apply for benefits but do not truly qualify.  And economists have researched this topic for years.

One recent paper by researchers at Columbia University, the Social Security Administration and the Congressional Budget Office (http://www.columbia.edu/~vw2112/papers/dissa_vwjsjm_final.pdf) finds that “younger rejected mail applicants to the Disability Insurance (DI) program exhibit substantial labor force attachment.  Similarly, a significant fraction of rejected applicants with low-mortality impairments such as back pain and mental health problems is employed.” 

In other words, there are a lot of people who apply for SSDI benefits, thus explicitly claiming that they have a work-ending disability, who return to work after being rejected.  Pretty clear evidence that they were not actually disabled, at least according to the SSDI definition.  But they applied for benefits anyway.  Maybe they really are hurt, maybe they really think they deserve the benefits.  But the fact that they can work after being rejected indicates that they did not suffer a work-ending disability. 

And as long as it remains the case that non-disabled people apply for disability benefits, the disability determination process will continue to be difficult, complex, long and extremely frustrating for everyone involved.  Those who suffer the most are those who truly are disabled.

In my last blog, I pointed out the inherent nature of uncertainties in climate projections, and the long list of reasons for particular uncertainties about the effects of anthropogenic greenhouse gas emissions on the change in future temperature levels, droughts, severe storms, sea level rise, and about measures of economic damages from any such event.  The range of possible outcomes is enormous, but I argued that the uncertainties are not a reason to wait and do more research before enacting legislation to reduce those emissions.  Indeed, the huge range of probability outcomes is a big reason to act now to reduce the possibility of such costly events.

In this blog, I want to expand that point to talk about the various kinds of uncertainties and what to do about them.  I just read an interesting blog by Keith Kloor that lists five kinds of reactions to uncertainty.  I will describe HIS five points, but what they bring to mind for me are the FIVE STAGES OF GRIEF (when a loved one dies, for example).  I’m sure you’ve heard these before:

1-Denial

2-Anger

3-Bargaining

4-Depression

5-Acceptance

Well, those approximately label his five reactions to uncertainty about climate change.  First, one could DENY the uncertainty, which might be done to try to further some political agenda.  Those who want environmental protection might say we KNOW that anthropogenic greenhouse gas emissions will cause significant global warming, and therefore we must act to prevent it.  That’s just wrong; we don’t KNOW that global warming will be significant and highly costly.

In fact, “uncertainty deniers” have done a great disservice to their own cause.  The claim that global warming is certain just gives the other side the opportunity to point out correctly that it’s NOT certain!  But that whole argument is irrelevant!  The relevant problem is that global warming MIGHT be significant and highly costly!

Second, one could react by trying to REDUCE the uncertainty, such as through herculean research efforts to make better predictions.  Research might well be worthwhile, and it might help reduce some of the uncertainties, but it will not reduce all of them, and it might introduce new uncertainties that we’ve not yet considered!

Third, one could try to SIMPLIFY the uncertainties, such as to explain in simple terms the complex scientific reasons for the inherent uncertainties listed in my previous blog.  It’s not wrong to try to explain complex uncertainties, and even to fit them into a finite set of categories, but the danger is that such simplification be taken as a replacement for consideration of all the complexities.   The problem is that simplification may in effect minimize those uncertainties.  Anyway, this kind of reaction is somewhat like bargaining: “maybe if we make up simple categories for these complex uncertainties then they might not seem so daunting.”

Actually, Kloor’s fourth reaction sounds even more like bargaining, when he says “Uncertainty detectives – well all scientists should work hard to understand, represent, and reason about uncertainty (. . .). The conflict is when political opponents seize on this uncertainty as an excuse for inaction.”  Now that is a cause for depression!

Anyway, of course, the fifth and final reaction to uncertainty is ACCEPTANCE: “include uncertainty information in rational decision support systems and policies.”  We need to know what is known, and what is unknown, to be able to make rational decisions as a society to adopt policies that can insure us against the worst possible outcomes.  We at least need to make the right tradeoffs between the costs of that insurance and the benefits of reducing those risks.  We need to undertake any available low cost measures to reduce fossil-fuel-fired electricity generation, to increase energy efficiency of vehicles and appliances, to increase alternative fuel use, to build water storage that can help deal with a possible increase in the number of droughts, and to build levees that can help deal with a possible increase in the number of severe storms.

Accepting the fact of uncertainty means giving up the idea of building in those protections because we know things will get worse.  Instead, it means building in those protections because things might get worse, and they might get a lot worse.

Nobody has any doubt that climate forecasts are uncertain.  They are uncertain with or without anthropogenic (human caused) effects of greenhouse gas emissions.  Then, when trying to gauge the effects of humans, we have to take the difference between the uncertain climate forecast with extra emissions and the uncertain forecast without extra emissions.  That only compounds the uncertainty!

Suppose for example that without our extra carbon dioxide and other greenhouse gas emissions, the temperature in 2050 is predicted to average 50°F plus or minus 5°.  And suppose the temperature with our current rate of emissions is predicted to average 52°F plus or minus 5°F.  Then the difference (the effect of emissions) is not just 52-50 = 2°F.  Rather, it could be anywhere between 57-45 = 12°F, at the high end, or 47-55 = -8°F, at the low end.  We just don’t know.

That simplified example is overstated.  But look at the figure from the IPCC Fourth Assessment Report’s “Summary for Policymakers”.  It shows a set of model simulations with a range of results anywhere from no global warming to about 4°C (which is about 7°F).   That is a lot of uncertainty, but that figure does not reflect all possible uncertainties.  Those include (but are not limited to): uncertainties about the amount of GHG emissions in the future, about the effect of those GHG emissions on ambient atmospheric concentrations, about the effect of ambient atmospheric concentrations on air temperature, about the effects of air temperature on ocean water temperature at different depths, about the feedback effects of ocean water temperature back on air temperature, the effects of all those changes on polar ice caps, the effects of polar ice caps on sea level rise, the effects of sea level rise on millions of miles of coastline around the world, and the effect of all those changes on economic damages.

Many have taken this inherent uncertainty as a reason not to act now, but instead to wait, to undertake more research, and to try to reduce that uncertainty.

That may be a natural initial reaction, but it is not a good one.  It assumes that uncertainty reduces the need to act, when in fact increases in uncertainty only increase the need to act!  That is not to say research is unwarranted, or that we have nothing more to learn. We can and should try to find out more and try to reduce uncertainties.  But a lot of that research may raise additional considerations and uncertainties!  Uncertainty is inherent to the problem and will never disappear, so waiting for resolution of the uncertainty means waiting forever and doing nothing forever.

Uncertainty itself is a problem we need to face, as it raises additional costs we can reduce.  A single hot summer or drought is a problem with which we have learned to cope.  But now we don’t even know whether we are facing that same level of heat and drought, or perhaps much more heat, reduced rainfall, extreme storms, huge loss of landmass, etc., etc., etc.  It is the unknown possibility of such loss that ought to make us act now to protect ourselves.

To the extent that anthropogenic GHG emissions raise uncertainties about future climate, the more we need policies that are resilient to those uncertainties: policies that increase our abilities to deal with drought, to make it possible to increase crop production with less rainfall, and to protect ourselves against the possibility of storms worse than Katrina.

Which brings us to the key distinction between adaptation and mitigation.  One way to protect ourselves is to adapt to droughts and storms, as just mentioned.  But another way to protect ourselves against those adverse possibilities is to start now to mitigate climate change by reducing GHG emissions.

A few days ago, AFL-CIO President Richard Trumka testified before the federal budget deficit commission.  In his remarks, he essentially argued (among other points) that we should try to grow our way out our problems.  Similarly, Edward Coyle, executive director of the Alliance for Retired Americans (an organization very closely affiliated with the union movement), objected to any discussion of raising the retirement age or reducing benefits.            

Sounds pretty good, right?  If we can just stimulate economic growth, we can avoid hard choices? 

Unfortunately, as with most “no pain” solutions to our nation’s fiscal problems, this one is too good to be true.  (In the name of bipartisanship, let me be clear that both Democrats and Republicans have their own version of the free lunch when it comes to Social Security – many free lunch Democrats argue we can grow out of the problem we have, and many free lunch Republicans believe that private accounts can solve the problem without benefit cuts.  Both are wrong – I will post about the flaws of the Republican form of free lunch at some other time.)

Let me be clear – growth is undoubtedly a good thing.  Of course I am pro growth.  Faster economic growth enlarges the economic pie, increases average wages, and thus provides more revenue for the same level of tax rates.  And there is no question that faster economic growth is a net positive for Social Security’s finances.

Unfortunately, faster growth is not sufficient to solve Social Security’s financial problems.  Let me point out two of the many reasons for this:

First, let’s remember that projections of Social Security’s long-term fiscal situation already assume that our economy will grow.  It is not as if Social Security’s trustees had not thought of this possibility.  So for growth to save us, it needs to be growth in excess of the baseline assumption.

Second, while it is true that faster growth and resultant higher wages increase payroll tax revenues to Social Security, this same wage growth also increases the benefits that Social Security must pay out in the future!  This is because the Social Security benefit formula is directly indexed to growth in the “average wage index.”  You may recall that the 2001 reform commission – and, in 2005, the Bush Administration itself – came out in favor of switching from a wage-indexed system to a price-indexed system.  Part of the rationale was to break this link and allow for us to get more of a fiscal “bang-for-the-buck” out of economic growth. 

There have been a lot of analyses to back up this analysis.  Of them all, the one that is most accessible to the non-PhD economist is probably the one written in 2003 by Rudolph Penner of the Urban Institute.  Hs conclusion: “Given the pending demographic pressures on the federal budget, we face a serious problem.  Increased growth cannot save us from breaking strong historical precedent.”  And that was back in 2003.  Sadly, the situation has gotten worse, not better …

So the short answer to the question posed in the title is “no.”

In my blog called “Buy Low and Sell High”, I talked about “arbitrage”:  with enough information about prices in different places, you can buy any commodity at a low price in one location and sell it simultaneously at a higher price in a different location.  If the price difference exceeds the transportation cost, you have a guaranteed profit.  Also, because of “tax arbitrage,” different jurisdictions may find it hard to charge different tax rates, if people can just buy their goods in a neighboring jurisdiction.   In the May 23 Champaign News-Gazette, “Gas Tax Proposal Will Have Obstacles” says that the City of Urbana plans to raise their gasoline tax by 2 cents per gallon.   It points out that the City of Champaign is very close, and many people commute back and forth and can easily buy gas in the other city. 

In that blog, I also pointed to a direct quote from Adam Smith (1776):  “High taxes, sometimes by diminishing the consumption of the taxed commodities, and sometimes by encouraging smuggling, frequently afford a smaller revenue to government than what might be drawn from more moderate taxes.”  In other words, Adam Smith knew that the increase in a tax rate might not lead to additional revenue. 

Here is another example of tax arbitrage.  A fascinating paper by David Merriman was just published in the American Economic Journal: Economic Policy, called “The Micro-Geography of Tax Avoidance: Evidence from Littered Cigarette Packs in Chicago.”  As of July 2007, the City of Chicago had a combined state and local tax of $3.66 per pack of cigarettes, while nearby Indiana had a state tax of only 55.5 cents per pack, and no local tax.  The Chicago tax rate also exceeds the rate in other nearby Illinois cities.  By purchasing cigarettes whenever they visit neighboring jurisdictions, residents of Chicago can easily engage in tax arbitrage.  If so, the tax is hard to enforce, and it may reduce total revenue to the city of Chicago.

But how would one proceed to measure this effect?   Nobody ever records the necessary data on transboundary purchases.  Stores in Indiana don’t ask where the buyer is from.  Nobody in Chicago checks whether your cigarettes are from out of town.

Merriman’s ingenious solution was to organize teams (of students, I presume) to collect littered cigarette packs in a representative random sample of areas in the City of Chicago.   They collected a total of 2,391 littered cigarette packs, of which 1,141 still had their tax stamps showing the place of purchase.  As it turns out, only 25 percent of littered packs in Chicago had a Chicago tax stamp!  Only 59 percent even had Illinois stamps, while 29% had Indiana stamps.  In particular Chicago locations that are very close to the Indiana border, 80 percent of littered packs had an Indiana tax stamp.

Even if nobody were really trying to avoid tax, however, plenty of people cross these borders every day and could randomly buy cigarettes in different places.  Even after subtracting that normal cross-border flow, however, Merriman finds statistically significant and large effects of tax avoidance. 

I’m not sure if these estimates can be used to calculate whether Chicago could cut its cigarette tax rate and actually raise more money.  But that would be an interesting additional calculation if possible – and Adam Smith would like to see it as well!

Yesterday I wrote about Social Security trust funds.  A reader with a terrific knowledge of Social Security history emailed me to point out an interesting fact that I thought was worth this short addendum.  I stated yesterday that the effect of the Greenspan commisison changes was to put it on a path to run surpluses for several decades.  That is true.  However, the wording of my next sentence (”these surpluses were to be ’saved’ in a ‘trust fund’ … that could then be drawn down once the demographic shift resulted in … deficits”) implies that these surpluses were part of an intentional effort to build up reserves that could be drawn down when the boomers retired.  In fact, as this knowledgable reader helpfully pointed out, that was not the logic used at the time.  He pointed out a quote from Bob Myers, who was the Executive Director of the Commission, who said “It was not intentional.  It has the effect of baby boomers paying for their own retirement … but it wasn’t a controlling element and it was not talked about.  The main thing that was talked about was how do we fix up the short-range problem.”  So the effect was as I described, but it was more of an accident than an intent, although the effect was as I described.  You can read more about this history by clicking here.   (Many thanks to the reader who pointed this out!)