Monday, January 23, 2017

Trade: Engine or Handmaiden of Growth?

How does international trade affect economic growth? This question has a pedigree. A half-century ago, it was common for economists to observe that the second half of the 19th century had seen a large wave of globalization and also a large wave of economic growth across many countries. Seemed as if the two might be connected! Back in 1970, the economist Irving Kravis challenged this consensus by drawing what has become a classic distinction in his article, "Trade as a Handmaiden of Growth: Similarities Between the Nineteenth and Twentieth Centuries," Economic Journal (December 1970, 80: 320, Dec. 1970, pp. 850-872).

Kravis's theme, which was controversial at the time and since, was that the connection from international trade to greater productivity did not arise primarily from an outright expansion of trade. He argued th at in the 19th century, some growth success stories expanded their trade while other did not. Instead, Kravis argued, economic growth was usually driven primarily by domestic factors like investment and education, and the presence of trade (not necessarily its expansion) played a smaller complementary role in translating these changes into economic growth. Thus, he argued that trade was not an "engine" of growth, but rather a "handmaiden" of growth.  For example, Kravis wrote:
"This evidence, it is argued below, does not support any simple generalisations about the dominant role of trade in the success stories of nineteenth- century growth. Export expansion did not serve in the nineteenth century to differentiate successful from unsuccessful countries. Growth where it occurred was mainly the consequence of favourable internal factors, and external demand represented an added stimulus which varied in importance from country to country and period to period. A more warranted metaphor that would be more generally applicable would be to describe trade expansion as a handmaiden of successful growth rather than as an autonomous engine of growth. ... 
"Perhaps the most important role played by trade is one that cannot be measured by trade statistics, viz., that a relatively open market enabled the growing country to find its areas of comparative advantage and to avoid the development of insulated, high-cost, inefficient sectors. In their direct impact, however, trade and capital movements were supplementary factors; they were handmaidens not engines of growth. The mainsprings of growth were internal; they must be sought in the land and the people, and in the system of social and economic organisation."
There's an ongoing argument in the research literature about how this argument applies to the 19th century evidence on globalization and growth. Those who are interested in the historical arguments can start by taking a look at "Trade as a Handmaiden of Growth: An Alternative View," by N. F. R. Crafts in the Economic Journal (September 1973, 83: 331, 875-884).

Here, I want to focus on the implications of Kravis's distinction at present. If trade is a main engine of growth, then it becomes important for trade to keep expanding as a share of GDP. If trade is a handmaiden of growth, then the presence of vigorous international trade is important, because it avoids what Kravis called "the development of insulated, high-cost, inefficient sectors," but continual expansions of trade don't matter nearly as much.

It's of course clear in the time since Kravis's 1970 essay, certain countries like Japan, Korea, and China have experienced a wave of economic growth that is related to their access to international markets. However, it's also clear that those countries have had very high levels of investment, as well as boosting the educational attainment and human capital of their population and being very willing to seek out and adopt new technologies. The US has had a large rise in its exports and imports in recent decades as a share of GDP, but economic growth in the US has fluctuated: fast in the 1960s, slow for much of the 1970s and 1980s, a surge roughly a decade long starting in the mid-1990s, and a growth slowdown since then. Up through about 2008, it was fair to say that the US economy has had the globalization, but not a corresponding large and sustained surge of productivity growth as a result. Since then, the US has experienced both a slowdown in trade growth and a slowdown in productivity, but of course this correlation doesn't prove a causal connection between the two.

For a modern overview at the relationship between international trade and productivity, Gary Clyde Hufbauer and Zhiyao (Lucy) Lu lay out the background in "Increased Trade: A Key to Improving Productivity" (October 2016, Peterson Institute for International Economics Policy Brief 16-15). They point out:"Global trade growth slowed abruptly after 2010,  following decades of expansion." They also offer a nice overview of recent developments in trade theory. (Those who would like more on developments in trade theory might begin with the four-paper symposium in the Spring 2012 issue of the Journal of Economic Perspectives.) Toward the end of the paper, they develop a rule-of-thumb for measuring the gains from trade.

To understand their approach, remember that it's certainly possible to have an expansion of trade with no particular gain in GDP.  Just consider an economy where both exports and imports rise by equal amounts, and the economy continues to produce the same amount. However, Hufbauer and Lu survey a number of studies about the effects of trade expansions and trade agreements on productivity. They argue that when trade expands, a certain percentage of that expansion represents efficiency gains from greater specialization in production and greater use of economies of scale. They write:
"[A] $1 billion increase in two-way trade increases potential GDP, through supply-side
efficiencies, by $240 million. ... Between 1990 and 2008, real US two-way trade in nonoil goods and services increased at an average rate of 5.86 percent a year. If two-way trade had increased at this pace after 2011, the real value of US two-way nonoil trade in 2014 would have been $308 billion greater than the observed value ($4.50 trillion versus $4.19 trillion). Based on the average dollar ratio of 0.24, the hypothetical increase in US two-way trade would have delivered a $74 billion increase in US GDP through supply-side efficiencies in 2014."
If you put this number in context, it's may not seem especially large. The US GPD was roughly $17 trillion in 2014, so an efficiency gain of $74 economy is less than half of 1%. To put it another way, say that size of US trade as a percentage of GDP increases by 0.4% per year over time. Then about one-fourth of that amount (the .24 figure from the Hufbauer and Lu estimates) represents an efficiency gain. By this quick-and-dirty measure, trade might add 0.1% per year to the US growth rate.

Even that estimate may be too high, because the effects of trade on productivity and growth are likely to differ quite substantially across countries. A boost in trade for a small economy that has been closed off from competition can help bring that economy into global supply chains, in a way that spurs growth through access to global technology and global markets. But the US is a very large economy with a reasonably competitive domestic market. For that kind of economy, an additional trade agreement is going to likely to have a much smaller effect. A number of studies (going back to Kravis and earlier) point out that trade may be of greater relative important for smaller economies, just as the North American Free Trade Agreement had a much larger positive effect for Mexico than the US economy.

But on the other side, the cautious reader may have noted that the Hufbauer-Lu estimate views trade from the "engine of growth" perspective: that is, the gains from trade come from expansions of trade, not from the "handmaiden of trade" effects like a competitive incentive for domestic firms to improve their efficiency and to focus on expanding into areas where their efficiency advantages are greatest, or the efficiency gains from trade that arise from learning more about other markets and technologies--even if trade itself isn't expanding.

It's also worth remembering that even seemingly small productivity gains, on the order of 0.1%, are cumulative over time. If several policies are all undertaken that can each raise growth by 0.1% per year, then after a few years the additional growth compounds to an economy that is noticeably bigger. A one-time gain of 0.1% of GDP in one year isn't a lot, but a permanent and ongoing gain of 0.1% of GDP every year is actually of meaningful if modest importance.

Overall, the world may have reached a pause in globalization, defined here as a rise in trade relative to GDP. In that sense, trade as an engine of growth has probably slowed. In addition, the gains for the US economy from signing additional trade agreements, given the enormous size and vast internal trade already present within the US economy, are not likely to be large--and certainly not large in the short-run. Long-run growth for the US economy is more likely to be based on investments in human capital, physicial capital, and technology. For smaller economies around the world, the possibility of greater participation in global markets can be considerably more important to their economic growth. For both large economies like the US and smaller economies around the world, the role of existing levels of international trade as a handmaiden of growth, providing competition and incentives and a check on industries that without such competition can become "insulated, high-cost, inefficient sectors," remains important.

Friday, January 20, 2017

What if Trump Skeptics, Like Me, Turn Out To Be Wrong?

I have been thinking back to the early 1980s, when I was graduating from college, and compiling a list of events that I never would have expected to see--but that have in fact happened. If you had asked me circa 1985:
  • I would have said that the Berlin Wall would not come down in my lifetime.
  • I would not have believed that the nations of Europe, and Germany in particular, would ever give up their traditional currencies for the euro. 
  • I would not  have believed that China would within a few decades become the largest economy in the world. 
  • I would not have believed that the Federal Reserve would take the federal funds interest rate down to near-zero and leave it there for seven full years. 
  • I would not have believed that the real estate developer who in 1983 opened Trump Tower in Manhattan and in 1984 opened the Harrah's casino at Trump Plaza in Atlantic City would ever become the President of the United States.
Just to be clear, I wouldn't have just said back in the first half of the 1980s that these events were merely unlikely. I would have viewed them as essentially unthinkable. Additions to this "I would not have believed" list are welcome: send them to For example, one friend contributed: "I would not have believed that the US presence in space would end up being spearheaded by the private sector."

It seems to me a useful mental discipline to admit when you are wrong--and especially when your errors demonstrate a substantial failure of imagination. Donald Trump was not my preferred or expected choice, either among those running for the Republican nomination or in the presidential election. I fear some of the potential consequences of his election. But I can clearly be wrong on major events, and I could be wrong about the effects of a President Trump, too. 

If a Trump presidency turns out badly in various ways, then Trump skeptics like me will certainly say so.  But if matters don't go wrong, then in fairness, then it seems to me that Trump skeptics should take a pledge to admit and acknowledge in a few years that at least some of our doubts and suspicions were incorrect--and indeed, we should be pleased that we were wrong.  Here's my version of that pledge on a few economic issues. 
  • If the US economy experiences a resurgence of manufacturing jobs, I will say so. 
  • If US economic growth surges to a 4% annual rate, I'll say so.
  • If the US economy does not actually retreat from foreign trade during four years of Trump presidency (which may well happen, given that globalization is driven by underlying economic forces, not just trade agreements), I will say so. 
  • If US carbon emissions fall during a Trump presidency (which may happen with the resurgence of cleaner-burning natural gas and the larger installed base of noncarbon energy sources), I will say so.
  • If the budget deficit does not explode in size during a Trump administration, despite all the promises for tax cuts and a huge boost in infrastructure spending, I will say so. 
  • If the Federal Reserve has maintained its traditional independence after 3-4 years, I will say so. 
  • If the number of Americans without health insurance is about the same in 3-4 years, or even lower, I will say so. 
These statements are not intended as predictions of what will or won't happen. My mother didn't raise any sons silly enough to make definite predictions about the future in print, and I have not tried to put a personal probability estimate on these outcomes. They are just possibilities. Of course, one can expand this list to include an array of other issues: what will happen in foreign policy hotspots from China and Latin America to the Middle East; patterns of economic and social inequality; fair treatment under the law for every single American; and many more.

On this Inauguration Day for President Donald Trump (and frankly, I still can't believe I am writing those words), I sincerely hope that I will turn out to be deeply incorrect about his readiness and fitness for office. I will try to observe what happens during a Trump administration clearly, without distortion through the prisms of my fears and disbeliefs, and without trying to justify my preexisting skepticism. After all, I've been wrong on big topics before. 

Tuesday, January 17, 2017

The Evolution of Medicaid and Health Financing Reform

Much of the public attention over the Patient Protection and Affordable Care Act of 2010 has focused on the "exchanges" through which households can receive subsidies for purchasing health insurance. But the main way in which the legislation expanded health insurance coverage is through altering eligibility for the Medicaid program. The Kaiser Family Foundation provides some details on how Medicaid works and how it has has changed in a  "Medicaid Pocket Primer" (January 3, 2017). As the report notes:
"The Medicaid program covers more than 70 million Americans, or 1 in 5, including many with complex and costly needs for care. ...  Medicaid covers a broad array of health services and limits enrollee out-of-pocket costs. The program is also the principal source of long-term care coverage for Americans. ... The Medicaid program finances over 16% of all personal health care spending in the U.S. ..
Before the Affordable Care Act (ACA), most low-income adults did not qualify for Medicaid because income eligibility for parents was very limited in most states – well below the federal poverty level (FPL) in most states ($11,880 in 2016) – and federal law excluded adults without dependent children from the program. These rules left many poor and low-income adults uninsured. As part of the broader framework the ACA established to cover uninsured Americans, the law expanded Medicaid to nonelderly adults with income up to 138% FPL – $16,394 for an individual in 2016. The ACA provided federal funding for the vast majority of the cost of the Medicaid expansion. ...

Under a 2012 Supreme Court ruling, the ACA Medicaid expansion is effectively optional for states. As of January 2017, 32 states including DC had expanded Medicaid and 19 states had not. ... Between Summer 2013, just prior to the ACA coverage expansions, and October 2016, Medicaid and CHIP enrollment rose by nearly 17 million. In 2015, an estimated 11 million enrollees were adults newly eligible for Medicaid under the ACA expansion and this number has likely grown as enrollment has continued to rise and additional states have expanded Medicaid. 
A few points are worth unpacking here. It's common in public discussions to refer to Medicaid as health insurance "for the poor," but that has never been quite correct. Adults without children have historically not been eligible for Medicaid, whether they were poor or not. Here are a couple of figures from the Kaiser report to illustrate the point. For example, the first bar shows that for the nonelderly below 100% of the federal poverty line, only 54% are covered by Medicaid, which as later bars show is breaks down into 76% of children and 40% of adults below the poverty line.

In fact, most Medicaid spending isn't aimed at the non-elderly poor. Here's another breakdown from Kaiser, showing that the disabled are 15% of Medicaid recipients, but receive 42% of all Medicaid spending, while the elderly are 9% of all Medicaid recipients, but receive 21% of all Medicaid spending (much of it for long-term care services).

The expansion of Medicaid enrollments has of course led to an increase in spending in what was already a very large program. The Kaiser report notes: "Total federal and state Medicaid spending was about $532 billion in FY 2015. Medicaid is the third-largest domestic program in the federal budget, after Social Security and Medicare, accounting for 9% of federal domestic spending in FY 2015. Medicaid is the second-largest item in state budgets, after elementary and secondary education, accounting for 18.7% of state general revenue spending and 28.2% of total state general revenue spending including federal funds to states, in 2015."

The Congressional Budget Office estimated last March that the expanded Medicaid enrollments in the Patient Protection and Affordable Care Act cost $67 billion per year.  One contributing factor is that the cost per patient of Medicaid expansion is turning out to be about 50% higher on a per capita than the earlier estimates, coming in at just over $6,300 per person, according to estimates last August from the Centers for Medicare and Medicaid Services.

Medicaid costs have climbed substantially over time as a share of GDP, from less than 0.5% of GDP when the program got underway in the late 1960s to more than 3% of GDP at present.

These patterns perhaps offer some useful context as the arguments over altering the Patient Protection and Affordable Care Act of 2010 gather momentum. As I've noted before, there's never been any secret that if the government was willing to spending tends of billions of dollars, it could expand health insurance coverage for millions of people.  Because I view the lack of health insurance coverage as a genuine problem, I'm fine with additional spending to expand Medicaid.

When President Obama addressed Congress about the pending health care legislation on September 9, 2009, he said: "So tonight, I return to speak to all of you about an issue that is central to that future -- and that is the issue of health care. I am not the first President to take up this cause, but I am determined to be the last." Even in September 2009, the notion that the US health care system could have a once-and-for-all fix was more of a rhetorical flourish than a practical reality. But given the form the actual legislation ended up taking, and given what has happened in the almost seven years since it was signed into law, the set of programs, regulations, and tax provisions affecting the US health care system clearly need some changes.  

Monday, January 16, 2017

Some Economics for Martin Luther King Jr. Day

On November 2, 1983, President Ronald Reagan signed the legislation establishing a federal holiday for the birthday of Martin Luther King Jr., to be celebrated each year on the third Monday in January. As the legislation that passed Congress said: "such holiday should serve as a time for Americans to reflect on the principles of racial equality and nonviolent social change espoused by Martin Luther King, Jr.." Of course, the case for racial equality stands fundamentally upon principles of justice, not economics. But here are five economics-related thoughts for the day excerpted from past posts, mostly but not entirely during the last year.

1) Inequalities of race and gender impose large economic costs on society as a whole, because one consequence of discrimination is that it hinders people in developing and using their talents. In "Equal Opportunity and Economic Growth" (August 20, 2012), I wrote:


A half-century ago, white men dominated the high-skilled occupations in the U.S. economy, while women and minority groups were often barely seen. Unless one holds the antediluvian belief that, say, 95% of all the people who are well-suited to become doctors or lawyers are white men, this situation was an obvious misallocation of social talents. Thus, one might predict that as other groups had more equal opportunities to participate, it would provide a boost to economic growth. Pete Klenow reports the results of some calculations about these connections in "The Allocation of Talent and U.S. Economic Growth," a Policy Brief for the Stanford Institute for Economic Policy Research.

Here's a table that illustrates some of the movement to greater equality of opportunity in the U.S. economy. White men are no longer 85% and more of the managers, doctors, and lawyers, as they were back in 1960. High skill occupation is defined in the table as "lawyers, doctors, engineers, scientists, architects, mathematicians and executives/managers." The share of white men working in these fields is up by about one-fourth. But the share of white women working in these occupations has more than tripled; of black men, more than quadrupled; of black women, more than octupled.

Moreover, wage gaps for those working in the same occupations have diminished as well. "Over the same time frame, wage gaps within occupations narrowed. Whereas working white women earned 58% less on average than white men in the same occupations in 1960, by 2008 they earned 26% less. Black men earned 38% less than white men in the typical occupation in 1960, but had closed the gap to 15% by 2008. For black women the gap fell from 88% in 1960 to 31% in 2008."

Much can be said about the causes behind these changes, but here, I want to focus on the effect on economic growth. For the purposes of developing a back-of-the-envelope estimate, Klenow builds up a model with some of these assumptions: "Each person possesses general ability (common to
all occupations) and ability specific to each occupation (and independent across occupations). All groups (men, women, blacks, whites) have the same distribution of abilities. Each young person knows how much discrimination they would face in any occupation, and the resulting wage they would get in each occupation. When young, people choose an occupation and decide how
much to augment their natural ability by investing in human capital specific to their chosen

With this framework, Klenow can then estimate how much of U.S. growth over the last 50 years or so can be traced to greater equality of opportunity, which encouraged many in women and minority groups who had the underlying ability to view it as worthwhile to make a greater investment in human capital.

"How much of overall growth in income per worker between 1960 and 2008 in the U.S. can be explained by women and African Americans investing more in human capital and working more in high-skill occupations? Our answer is 15% to 20% ... White men arguably lost around 5% of their earnings, as a result, because they moved into lower skilled occupations than they otherwise would have. But their losses were swamped by the income gains reaped by women and blacks."

At least to me, it is remarkable to consider that 1/6 or 1/5 of total U.S. growth in income per worker may be due to greater economic opportunity. In short, reducing discriminatory barriers isn't just about justice and fairness to individuals; it's also about a stronger U.S. economy that makes better use of the underlying talents of all its members.


2) An "audit study" of housing discrimination involves finding pairs of people, giving them similar characteristics (job history, income, married/unmarried, parents/not parents) and sending them off to buy or rent a place to live. In "Audit Studies and Housing Discrimination" (September 21, 2016), I wrote in part:

Cityscape magazine, published by the US Department of Housing and Urban Development three times per year, has a nine-paper symposium on "Housing Discrimination Today" in the third issue of 2015. The lead article by Sun Jung Oh and John Yinger asks: "What Have We Learned From Paired Testing in Housing Markets?" (17: 3, pp. 15-59). ...

There have been four large national-level paired testing studies of housing discrimination in the US in the last 40 years. "The largest paired-testing studies in the United States are the Housing Market Practices Survey (HMPS) in 1977 and the three Housing Discrimination Studies (HDS1989, HDS2000, and HDS2012) sponsored by the U.S. Department of Housing and Urban Development (HUD)." Each of the studies were spread over several dozen cities. The first three involved about 3,000-4,000 tests; the 2012 study involved more than 8,000 tests. The appendix also lists another 21 studies done in recent decades.

Overall, the findings from the 2012 study find ongoing discrimination against blacks in rental and sales markets for housing. For Hispanics, there appears to be discrimination in rental markets, but not in sales markets. Here's a chart summarizing a number of findings, which also gives a sense of the kind of information collected in these studies.

However, the extent of housing discrimination in 2012 has diminished from previous national-level studies. Oh and Yinger write (citations omitted): "In 1977, Black homeseekers were frequently denied access to advertised units that were available to equally qualified White homeseekers. For instance, one in three Black renters and one in every five Black homebuyers were told that there were no homes available in 1977. In 2012, however, minority renters or homebuyers who called to inquire about advertised homes or apartments were rarely denied appointments that their White counterparts were able to make.

3) Many of the communities that suffer the most from crime are also the communities where the law-abiding and the law-breakers both experience a heavy law enforcement presence, and where large numbers of young men end up being incarcerated. Here are some slices of my discussion from "Inequalities of Crime Victimization and Criminal Justice" (May 20, 2016):

And law-abiding people in some communities, many of them predominantly low-income and African-American, can end up facing an emotionally crucifying choice. One one side, crime rates in their community are high, which is a terrible and sometimes tragic and fatal burden on everyday life. On the other side, they are watching a large share of their community, mainly men, becoming involved with the criminal justice system through fines, probation, fines, or incarceration. Although those who are convicted of crimes are the ones who officially bear the costs, in fact the costs when someone needs to pay fines, or can't earn much or any income, or can only be visited by making a trip to a correctional facility are also shared with families, mothers, and children. Magnus Lofstrom and Steven Raphael explore these questions of "Crime, the Criminal Justice System, and Socioeconomic Inequality" in the Spring 2016 issue of the Journal of Economic Perspectives. ...

It's well-known that rates of violent and property crime have fallen substantially in the US in the last 25 years or so. What is less well-recognized is that the biggest reductions in crime have happened in the often predominantly low-income and African-American communities that were most plagued by crime. Loftrom and Raphael look at crime rates across cities with lower and higher rates of poverty in 1990 and 2008:
"However, the inequality between cities with the highest and lower poverty rates narrows considerably over this 18-year period. Here we observe a narrowing of both the ratio of crime rates as well as the absolute difference. Expressed as a ratio, the 1990 violent crime rate among the cities in the top poverty decile was 15.8 times the rate for the cities in the lowest poverty decile. By 2008, the ratio falls to 11.9. When expressed in levels, in 1990 the violent crime rate in the cities in the upper decile for poverty rates exceeds the violent crime rate in cities in the lowest decile for poverty rates by 1,860 incidents per 100,000. By 2008, the absolute difference in violent crime rates shrinks to 941 per 100,000. We see comparable narrowing in the differences between poorer and less-poor cities in property crime rates. ... "
It remains true that one of the common penalties for being poor in the United States is that you are more likely to live in a neighborhood with a much higher crime rate. But as overall rates of crime have fallen, the inequality of greater vulnerability to crime has diminished.

On the other side of the crime-and-punishment ledger, low-income and African-American men are more likely to end up in the criminal justice system. Lofstrom and Raphael give sources and studies for the statistics: "[N]nearly one-third of black males born in 2001 will serve prison time at some point in their lives. The comparable figure for Hispanic men is 17 percent ... [F]or African-American men born between 1965 and 1969, 20.5 percent had been to prison by 1999. The comparable figures were 30.2 percent for black men without a college degree and approximately 59 percent for black men without a high school degree."

I'm not someone who sympathizes with or romanticizes those who commit crimes. But economics is about tradeoffs, and imposing costs on those who commit crimes has tradeoffs for the rest of society, too. For example, the cost to taxpayers is on the order of $350 billion per year, which in 2010 broke down as "$113 billion on police, $81 billion on corrections, $76 billion in expenditure by various federal agencies, and $84 billion devoted to combating drug trafficking." The question of whether those costs should be higher or lower, or reallocated between these categories, is a worthy one for economists. ...  Lofstrom and Raphael conclude:

"Many of the same low-income predominantly African American communities have disproportionately experienced both the welcome reduction in inequality for crime victims and the less-welcome rise in inequality due to changes in criminal justice sanctioning. While it is tempting to consider whether these two changes in inequality can be weighed and balanced against each other, it seems to us that this temptation should be resisted on both theoretical and practical grounds. On theoretical grounds, the case for reducing inequality of any type is always rooted in claims about fairness and justice. In some situations, several different claims about inequality can be combined into a single scale—for example, when such claims can be monetized or measured in terms of income. But the inequality of the suffering of crime victims is fundamentally different from the inequality of disproportionate criminal justice sanctioning, and cannot be compared on the same scale. In practical terms, while higher rates of incarceration and other criminal justice sanctions may have had some effect in reducing crime back in the 1970s and through the 1980s, there is little evidence to believe that the higher rates have caused the reduction in crime in the last two decades. Thus, it is reasonable to pursue multiple policy goals, both seeking additional reductions in crime and in the continuing inequality of crime victimization and simultaneously seeking to reduce inequality of criminal justice sanctioning. If such policies are carried out sensibly, both kinds of inequality can be reduced without a meaningful tradeoff arising between them."

While accusations of police brutality are often the flashpoint for public protests over the criminal justice system, my own suspicion is that some of the anger and despair focused on the police is because they are the visible front line of the criminal justice system. It would be interesting to watch the dynamics if protests of similar intensity were aimed at legislators who pass a cavalcade of seemingly small fines, which when imposed by judges add up to an insuperable burden for low-income families. Or if the protests were aimed at legislators, judges, and parole boards who make decisions about length of incarceration. Or if the protests were aimed at prisons and correctional officers. My own preference for the criminal justice system (for example, here and here) would be to rebalance the nation's criminal justice spending, with more going to police and less coming in fines, and the offsetting funding to come from reducing the sky-high levels of US incarceration. The broad idea is to spend more on tamping down the chance that crime will occur or escalate in the first place, while spending less on years of severe punishments after the crime has already happened.

4) There is a popular movement toward "ban-the-box," which prohibits employers from including a question on their application forms that asks if you have a criminal history. The hope behind such a rule is that it will improve educational opportunities for African-Americans, who are statistically more likely to have a criminal history. But several empirical studies suggest that it has the opposite effect, as I explained in "How Ban the Box Reduces Job Opportunities for African-Americans" (September 16, 2016): 


The study is called "Ban the Box, Criminal Records, and Statistical Discrimination: A Field Experiment," by Amanda Agan and Sonja Starr (Princeton University International Relations Section, Working Paper #5998, July 2016). ... As Agan and Starr write at the start of the paper (citations omitted):
In an effort to reduce barriers to employment for people with criminal records, more than
100 jurisdictions and 23 states have passed “Ban-the-Box” (BTB) policies. Although the details vary, these policies all prohibit employers from asking about criminal history on the initial job application and in job interviews; employers may still conduct criminal background checks, but only at or near the end of the employment process. Most BTB policies apply to public employers only, but seven states (including New Jersey) and a number of cities (including New York City) have now also extended these restrictions to private employers. These laws seek to increase employment opportunities for people with criminal records. They are often also presented as a strategy for reducing unemployment among black men, who in recent years have faced unemployment rates approximately double the national average ... 
Agan and Starr carried out an experiment. They sent out about 15,000 fictitious online job applications to entry-level positions in New Jersey and New York city, both before and after the "ban-the-box" policy went into effect. The resumes were set up in pairs, so that they were largely the same resume except for a difference in race; in particular, out of each pair, one job applicant could be identified as white and one as black. In addition, some of the pairs of hypothetical applicants checked "the box" early on, while others did not; some had a high school diploma, or a GED high-equivalency, or neither; some had a gap in their job history, while others did not.

The study found that whites with the same credentials are more likely to get a call-back than blacks: as they write, "white applicants overall received about 23% more callbacks compared to similar black applicants." Before "ban-the-box" went into effect, admitting to a criminal record definitely made it harder to be hired: that is, "among employers that asked about criminal convictions in the pre-period, the effect of having a felony conviction is also significant and large: applicants without a felony
conviction are 62% ... more likely to be called back than those with a conviction, averaged across races ..."

However, when ban-the-box (BTB) was enacted, the black-white gap in the chances of being called back got larger, not smaller. "Our estimates of BTB’s effects on callback rates imply that BTB substantially increases racial disparities in employer callbacks. We find that BTB expands the black-white gap by about 4 percentage points, multiplying the gap at affected businesses by a factor of about six. In our main specification, before BTB, white applicants to BTB-affected employers received 7% more callbacks than similar black applicants, but after BTB this gap grew to 45% ..."

The authors suggest that what economists call "statistical discrimination" is a possible explanation for these findings. ... Consider an employer who is both mildly biased against blacks, but also would strongly prefer not to hire someone who has a criminal record. If that employer has information on whether someone has a criminal record, they will continue to be biased against blacks. But if this employer is banned from collecting information on criminal record, they will tend to act on the statistical knowledge that blacks are more likely to have a criminal record than whites. As a result, blacks without a criminal record will have a lower chance of a job callback, and whites with a criminal record will have a higher chance of a job callback. ...

Thanks to Catherine Rampell for pointing out to me that there's another recent empirical study of ban-the-box, different methods, but similar results. The study is "Does "Ban the Box" Help or Hurt Low-Skilled Workers? Statistical Discrimination and Employment Outcomes When Criminal Histories are Hidden," by Jennifer L. Doleac and Benjamin Hansen, published as NBER Working Paper No. 22469 (July 2016). ... Thanks to Stan Veuger for pointing out yet another recent working paper on this subject, which uses a different approach and emphasizes a different set of tradeoffs. In "No Woman No Crime: Ban the Box, Employment, and Upskilling," Daniel Shoag and Stan Veuger look at employment with a focus on the outcome of ban-the-box an employment rates of those living in high-crime neighborhoods.

5) Is it possible to draw a reasonably clear line between equality of opportunity and equality of result? Here are some excerpts from my thoughts on Equality of Opportunity and Equality of Result (June 17, 2015):


In discussions about inequality of income or wealth, it's common to hear an argument along the following lines: "I'm not much bothered by inequality of results, as long as there is fairly good equality of opportunity."

As a quick example of this distinction, consider two siblings of the same gender that grow up in the same family, attend the same schools and colleges, and get similar jobs. However, one sibling saves money for retirement, while the other does not. When the two of them reach retirement, one sibling can afford around-the-world cruises and extensive pampering of the grandchildren, while the other sibling can afford the early-bird discount diner buffet line. This inequality of after-retirement results between the two siblings doesn't seem especially bothersome, because of the earlier equality of opportunities.

However, the notion that the inequality resulting from different opportunities or discrimination can be more-or-less separated from the inequality that results from choices and effort, while appealing at an intuitive level, turns out to quite difficult in practice. Ravi Kanbur Adam Wagstaf discuss the issues in "How Useful Is Inequality of Opportunity as a Policy Construct?" in World Bank Policy Research Working Paper 6980 (July 2014). The authors have also written a recent short summary/overview of the arguments. As a starting point, they write:

In policy and political discourse, “equality of opportunity” is the new motherhood and apple pie. It is often contrasted with equality of outcomes, with the latter coming off worse. Equality of outcomes is seen variously as Utopian, as infeasible, as detrimental to incentives, and even as inequitable if outcomes are the result of differing efforts. Equality of opportunity, on the other hand, is interchangeable with phrases such as ‘leveling the playing field’, ‘giving everybody an equal start’ and ‘making the most of inherent talents.’ In its strongest form, the position is that equality of outcomes should be irrelevant to policy; what matters is equality of opportunity. ... However, attempts to quantify and apply the concept of equality of opportunity in a policy context have also revealed a host of problems of a conceptual and empirical nature, problems which may in the end even question the practical usefulness of the concept.

My sense that their argument can be divided into two parts. One problem is that it's not easy to divide up the inequalities that are observed in society into one portion based on differences in opportunity, which should be rooted in the circumstances in which people find themselves through no decision or fault of their own, and another portion based on the choices or efforts that people make. The other problem is that moral intuition in some cases suggests an aggressive role for acting against unequal opportunities, and other cases where the moral intuition is not as strong: for example, the argument for fighting race and gender discrimination in support of equality of opportunity seems considerably stronger than the argument for seeking to offset most differences in genetic talent as a way of ensuring equal opportunity. ...

The difficult bottom line here is that seeking to draw a distinction between equality of opportunity and equality of results hides a deeper question: What sources of unequal results should a society regard as acceptable or justified, and what sources of unequal results should we regard as unacceptable or unjustified? It's easy to claim that such a distinction exists, but knowing in practical terms where it can be difficult.

In a 1965 speech, President Lyndon Johnson discussed the importance of true equality of opportunity in a famous passage:
But freedom is not enough. You do not wipe away the scars of centuries by saying: Now you are free to go where you want, and do as you desire, and choose the leaders you please. You do not take a person who, for years, has been hobbled by chains and liberate him, bring him up to the starting line of a race and then say, "you are free to compete with all the others," and still justly believe that you have been completely fair. Thus it is not enough just to open the gates of opportunity. All our citizens must have the ability to walk through those gates.Johnson's comment contains a deep truth, but the poetic phrasing about starting lines of races and walking through gates of opportunity offers a hint that practical difficulties are being sidestepped.
For example, it is straightforward to argue that children should have at least some minimum level of opportunity, a feeling which is expressed by laws requiring compulsory and taxpayer-funded schooling and public health measures like vaccinations. But beyond that minimum, the extent to which society should intervene with parents or seek to counterbalance or offset parental decisions about raising children can become quite controversial. It is straightforward to argue that racial and ethnic discrimination should be banned. But beyond the essential step of banning explicit discrimination in employment or housing or public services, the extent to which society should act to offset the results of past discrimination becomes controversial, too. ...

Overall, it seems that the distinction between equality of opportunity and equality of result can be the starting point for some minimum level of public policy to reduce certain causes of unequal outcomes. But given the analytical problem with separating why unequal results occur, the equality of opportunity/equality of result distinction is often not much help in resolving how aggressive such inequality-reducing policies should be.

Saturday, January 14, 2017

Want to Watch Some Economists Talk?

The annual meetings of the American Economic Association were held in Chicago from January 6-8. The AEA offers webcasts of a number of popular sessions, often panel discussions or prominent lectures in which the discussion should be reasonably accessible to nonspecialists. A sampling of these webcasts is below; a full list of the available webcasts from the meetings is here. If you would like to hear actual economists of differing views talk about issues, with enough time for sentences and even paragraphs rather than just soundbites, these webcasts offer a nice starting point.

Brexit: Six Months Later (Panel Discussion)
Presiding: Olivier Blanchard
Jonathan Portes
Andrew Lilico
Karl Whelan
View Webcast

Nobels on Where is the World Economy Headed?
Presiding: Dominick Salvatore
Where in the World Is the World Headed? Angus Deaton
Seeking Political Keys for Economic Growth Roger Myerson
How the Left and Right Are Failing the West Edmund Phelps
Economic Risks Associated with Deep Change in Technology Robert J. Shiller
New Divisions in the World Economy Joseph E. Stiglitz
View Webcast

AEA Richard T. Ely Lecture: The Economist as Plumber: Large Scale Experiments to Inform the Details of Policy Making
Esther Duflo, introduced by Alvin E. Roth
View Webcast

Economic Issues Facing the New President (Panel Discussion)
Presiding: Greg Mankiw
Jason Furman
Glenn Hubbard
Alan Krueger
John Taylor
View Webcast

AEA Presidential Address - Narrative Economics
Robert J. Shiller, introduced by Alvin E. Roth
View Webcast

Friday, January 13, 2017

A New Era of Price Discrimination?

"Price discrimination" has a specific technical meaning for economists. It's not about sellers charging more to certain groups because of biased attitudes about gender, race/ethnicity, religion, or sexual orientation. Instead, it's about setting up a varying set of prices in order to charge more to those who are willing to pay more--unlike the standard situation in a plain vanilla market in which everyone pays the same price.

There are lots of examples of price discrimination. When a movie is first released, the ticket prices are typically higher in "first-run" theaters than when the movie arrives at "second-run" theaters a few months later. Books are often released first in more-expensive hard-cover editions, and later in less-expensive paperbacks. Those who aren't sure about going out for dinner are enticed by happy hour and early-bird specials, while those willing to pay more arrive later in the evening. There are discounts for students or senior citizens. There are volume discounts for buying a larger quantity of a good. Such arrangements often seem potentially beneficial to both buyers and sellers.

But there's one more kind of price discrimination called "personalized pricing," in which prices would vary across individuals so that everyone would be charged as much as they were willing to pay. This seems more problematic, and a combination of big data and online retail may be bringing it our way. Ariel Ezrachi and Maurice E. Stucke write about "The rise of behavioural discrimination," in the European Competition Law Review (2016, 12: 485-492; not freely available online). They also refer to an Executive Office of the President report from February 2015, "Big Data and Differential Pricing." That report sets up the issue in this way:

"Economics textbooks usually define three types of differential pricing. Personalized pricing, or first-degree price discrimination, occurs when a seller charges a different price to every buyer. Individually negotiated prices, such as those charged by a car dealer, are an example of personalized pricing. Quantity discounts, or second-degree price discrimination, occur when the per-unit price falls with the amount purchased, as with popcorn at the movie theater. Finally, third-degree price discrimination occurs when sellers charge different prices to different demographic groups, as with discounts for senior citizens.
Big data has lowered the costs of collecting customer-level information, making it easier for sellers to identify new customer segments and to target those populations with customized marketing and pricing plans. The increased availability of behavioral data has also encouraged a shift from third-degree price discrimination based on broad demographic categories towards personalized pricing. Nevertheless, differential pricing still presents several practical challenges. First, sellers must figure out what customers are willing to pay. This can be a complex problem, even for companies with lots of data and computing power. A second challenge is competition, which limits a company’s ability to raise prices, even if it knows that one customer might be willing to pay more than another. Third, companies need to prevent resale by customers seeking to exploit price differences. And finally, if a company does succeed in charging personalized prices, it must be careful not to alienate customers who may view this pricing tactic as inherently unfair. ...
Ultimately, whether differential pricing helps or harms the average consumer depends on how and where it is used. In a competitive market with transparent pricing, the benefits are likely to outweigh the costs.  ...  Ultimately, differential pricing seems most likely to be harmful when implemented through complex or opaque pricing schemes designed to screen out unsophisticated buyers. For example, companies may obfuscate by bundling a low product price with costly warranties or shipping fees, using “bait and switch” techniques to attract unwary customers with low advertised prices and then upselling them on different merchandise, or burying important details in the small print of complex contracts.When these tactics work, the economic intuition that differential pricing allows firms to serve more price-sensitive customers at a lower price-point may even be overturned. If price-sensitive customers also tend to be less experienced, or less knowledgeable about potential pitfalls, they might more readily accept offers that appear fine on the surface but are actually full of hidden charges. ...."
Ezrachi and Stucke point out a number of ways in which these issues are becoming a practical reality. The collection and interconnection of big data from a wide variety of sources creates the possibility that when you shop on-line, the seller may already know quite a lot about you. They write:

"As the volume, variety and value of personal data increases, self-learning pricing algorithms can use the data collected on you and other people to identify subgroups of like-minded, like-price-sensitive individuals, who share common biases and levels of willpower. Pricing algorithms can use data on how other people within your grouping react, to predict how you will likely react under similar circumstances. This then enables the self-learning algorithm to more accurately approximate the user's reservation price, observe behaviour, and adjust. The more time we spend online--chatting, surfing, and purchasing--the more times the algorithm can observe what you and others within your grouping do under various circumstances; the more experiments it can run; the more it can learn through trial and error what your group's reservation price is under different situations; and, the more it can recalibrate and refine (including shifting you to another group). 
"To better train their algorithms and categorize even smaller groups of individuals, firms will need personal data. Among other things, this trend will accelerate the "Internet of Things", as firms compete to collect data on consumers' activities at home, work, and outside. Smart appliances, cars, utensils, and watches can help firms refine their consumer profiles and gain a competitive edge. Thus in making use of our demographics, physical location (via our phones), browser and search history, friends and links on social networks, and online reviews and blog posts, firms can target us with personalised advertisements with ever increasing proficiency. Also, at the point of sale, the categorisation can help sellers approximate our price sensitivity."
It used to be said that when you go to a website, you are like a person with a name-tag at a convention: that is, you could be identified, but others didn't necessarily know much about you. But in the future, when you go to a website, certain sellers at least will already know a great deal about you. With this information, the seller will be able to customize your retail experience by manipulating the information presented about products, choices, prices, and deals in ways that makes someone with your specific characteristics more likely to buy and to pay higher prices.

This could be done in literally dozens of ways. One example from Ezrachi and Stucky is that the first item presented in an online list of possibilities will both be a decoy designed with your characteristics in mind: it will also be higher-priced, and perhaps lacking in some features.  When you scroll down the list, you will find other items that have lower prices or more features. Compared to the decoy item, these look like good deals. A standard example in regular retailing is that many restaurants report that the second most-expensive bottle of wine and the second least-expensive bottle of wine are among their top seller, because those who want to splurge can feel they are being a little thrifty, and those who want inexpensive can feel they aren't being totally cheap. "So we may have originally intended to purchase a cheaper item, but chose a more expensive item with perhaps a few more attributes, as it was relatively more attractive than the personalised decoy option."

Another option is "price-steering," where a website makes it easier to find more expensive options. Or firms can make strategic use of complexity: "To better discriminate, companies can take advantage of consumers' difficulty in processing many complex options. Companies may deliberately increase the complexity by adding price and quality parameters, with the intent to facilitate consumer
error or bias and manipulate consumer demand to their advantage. By increasing their products' complexity, firms can also make it difficult to appraise quality and compare products, increase the consumers' search and evaluation costs, and nudge consumers to rely on basic signalling that benefits the firms. Once the customer is snagged, the complexity in contract terms can increase
the customers' switching costs and increase the likelihood of customers retaining the personalised default option.  This enables firms to inch closer to perfect behavioural discrimination."

Notice that none of these strategies involve the seller actually lying. In fact, one can easily think of circumstances where these options could benefit consumers, by providing them with the selection of products and information that they actually find most attractive. But it's also easy to think of ways in which people can be manipulated. Ezrachi and Stucke write:
The road to near-perfect behavioural discrimination will be paved with personalised coupons and promotions: the less price-sensitive online customers may not care as much if others are getting promotional codes, coupons, and so on, as long as the list price does not increase. Online sellers will increasingly offer consumers with a lower reservation price a timely coupon-ostensibly for being a valued customer, a new customer, a returning customer, or a customer who won the discount. The coupon may appear randomly assigned, but only customers with a lower reservation price are targeted. Indeed, the price discrimination can happen on other, less salient aspects of the purchase. Retailers can offer the same price, but provide greater discounts on shipping (or faster delivery), offer complimentary customer service, or better warranty terms to attract customers with lower reservation prices, greater willpower, or more outside options.
In the brave new world of big data and online purchases. buyers really do need to be wary. And one suspects that the Federal Trade Commission and other consumer protection agencies are going to become active participants in determining what tools sellers can use.

Monday, January 9, 2017

Narrative Economics and the Laffer Curve

Robert Shiller delivered the Presidential Address for the American Economic Association on the subject of "Narrative Economics" in Chicago on January 7, 2017. A preliminary version of the underlying paper, together with slides from the presentation, is available here.

Shiller's broad point was that the key distinguishing trait of human beings may be that we  organize what we know in the form of stories.  He argues:
"Some have suggested that it is stories that most distinguish us from animals, and even that our species be called Homo narrans (Fisher 1984) or Homo narrator (Gould 1994) or Homo narrativus (Ferrand and Weil 2001) depending on whose Latin we use.  Might this be a more accurate description than Homo sapiens, i.e., wise man? Or might we say "narrative is intelligence" (Lo, 2007), with all of its limitations? It is more flattering to think of ourselves as Homo sapiens, but not necessarily more accurate."
Shiller goes on to make a case that narratives play a role in economic activity: for example, the way people act during the steep recession of 1920-21 and the Great Depression, as well as in the Great Recession and the most recent election. To me, one of his themes is that economist should seek to bring the narratives of these times that economic actors were telling themselves into their actual analysis by applying epidemiology models to examine actual spread of narratives, rather than bewailing narratives as a sort of unfair complication for the purity of our economic models.

Near the start, Shiller offers the Laffer Curve as an example of a narrative that had some lasting force. For those not familiar with the story, here's how Shiller tells it (footnotes omitted):
Let us consider as an example the narrative epidemic associated with the Laffer curve, a diagram created by economist Arthur Laffer ... The story of the Laffer curve did not go viral in 1974, the reputed date when Laffer first introduced it. Its contagion is explained by a literary innovation that was first published in a 1978 article in National Affairs by Jude Wanniski, an editorial writer for the Wall Street Journal. Wanniski wrote the colorful story about Laffer sharing a steak dinner at the Two Continents [restaurant] in Washington D.C. in 1974 with top White House powers Dick Cheney [at the time, a Deputy Assistant to President Ford, later to be Vice President] and Donald Rumsfeld (at the time Chief of Staff to President Ford, later to be Secretary of Defense]. Laffer drew his curve on a napkin at the restaurant table.  When news about the "curve drawn on a napkin" came out, with Wanniski's help, the story surprisingly went viral, so much that it is now commemorated. A napkin with the Laffer curve can be seen at the National Museum of American History ... 
Why did this story go viral? Laffer himself said after the Wanniski story exploded that he himself could not remember the event, which had taken place four years earlier. But Wanniski was a journalist who sensed that he had the elements of a good story. The key idea as Wanniski presented it is, indeed, punchy: At a zero-percent tax rate, the government collects no revenue. At a 100% tax rate the government would also collect no revenue, because people will not work if all the income is taken. Between the two extremes, the curve, relating tax revenue to tax rate, must have an inverted U shape. ...
Here is a notion of economic efficiency easy enough for anyone to understand. Wanniski suggested, without any data, that we are on the inefficient side of the Laffer curve. Laffer's genius was in narratives, not data collection. The drawing of the Laffer curve seems to suggest that cutting tax rates would produce a huge windfall in national  income. To most quantitatively-inclined people unfamiliar with economics, this explanation of economic inefficiency was a striking concept, contagious enough to go viral, even though economists, even though economists protested that we are not actually on the inefficient side of the Laffer Curve (Mirowski 1982). It is apparently impossible to capture why it is doubtful that we are on the inefficient side of the Laffer curve in so punch a manner that it has the ability to stifle the epidemic. Years later Laffer did refer broadly to the apparent effects of historic tax cuts (Laffer 2004); but in 1978 the narrative dominated. To tell the story really well one must set the scene at the fancy restaurant, with powerful Washington people and the napkin.
Here an image of what mus be one of history's best-known napkins from the National Museum of American History, which reports that the exhibit was "made" on September 14, 1974, and measures 38.1 cm x 38.1 cm x .3175 cm, and was a gift from Patricia Koyce Wanniski:

Did Laffer really pull out a pen and start writing on a cloth napkin at a fancy restaurant, so that Jude Wanniski could take the napkin away with him? The website of the Laffer Center at the Pacific Research Institute describes it this way:
"As to Wanniski’s recollection of the story, Dr. Laffer has said that he cannot remember the details, but he does recall that the restaurant where they ate used cloth napkins and his mother had taught him not to desecrate nice things. He notes, however, that it could well be true because he used the so-called Laffer Curve all the time in classroom lectures and to anyone else who would listen." 
In the mid-1980s, when I was working as an editorial writer for the San Jose Mercury News in California, I interviewed Laffer when he was running for a US Senate seat.  He was energy personified and talked a blue streak, and I can easily imagine him writing on cloth napkins in a restaurant. When remembering the event 40 years later in 2014, Dick Cheney said:
It was late afternoon, sort of the-end-of-the-day kind of thing. As I recall, it was a round table. I remember a white tablecloth and white linen napkins because that’s what [Laffer] drew the curve on. It was just one of those events that stuck in my mind, because it’s not every day you see somebody whip out a Sharpie and mark up the cloth napkin at the dinner table. I remember it well, because I can’t recall anybody else drawing on a cloth napkin.
The point of Shiller's talk is that while a homo sapiens discussion of the empirical evidence behind the Laffer curve can be interesting in its own way, understanding the political and cultural impulse behind tax-cutting from the late 1970s up to the present requires genuine intellectual opennees to a homo narrativus explanation--that is, an understanding of what narratives have force at certain times, how such narratives come into being, why the narratives are powerful, and how the narratives affect various forms of economic behavior.

My own sense is that homo sapiens can be a slippery character in drawing conclusions. Homo sapiens likes to protest that all conclusions come from a dispassionate consideration of the evidence. But again and again, you will observe that when a certain homo sapiens agrees with the main thrust of a certain narrative, the supposedly dispassionate consideration of evidence involves compiling every factoid and theory in support, as well as denigrating those who believe otherwise as liars and fools; conversely, when a different homo sapiens disagrees with the main thrust of certain narrative, the supposedly dispassionate consideration of the evidence involves compiling every factoid and theory in opposition, and again denigrating those who believe otherwise as liars and fools. Homo sapiens often brandishes facts and theories as a nearly transparent cover for the homo narrativus within.