Thursday, August 8, 2019

Place-based economic policies (pt. 2 of 3): How effective are tax incentives for investments in low-income communities

In this post, I discuss tax incentives for investments in low-income communities. In the U.S., these incentives often have bipartisan support. Democratic presidential contender and Senator Cory Booker and Republican Senator Tim Scott were the primary architects of the provision in the recent Trump tax bill (Tax Cuts and Jobs Act or TCJA) that allows investors to forego capital gains taxes on long-term investments made in low-income Census tracts that they designate as "opportunity zones". These incentives are by no means recent or unique. This policy paper from the Minnesota House of Representatives provides a concise summary of the history and implementation of enterprise zones in the U.S. and evidence from the literature - through 2005 - on their effectiveness.

In theory, economic barriers prevent qualifying low-income communities from reaching their full potential (lack of transportation, lack of access to capital, lack of skilled labor, social problems, environmental problems) however tax incentives can outweigh the costs to investors of investing in these areas. As stated aptly in the paper: "In the language of economics, the first, best solution is to find a subsidy that equates the marginal social benefits with the marginal social costs to doing business within the zone. Deciding upon the value of the social benefits is a difficult task, let alone determining how much of a subsidy is needed to attract the needed number of businesses."

These decisions are made by our elected officials (and therefore indirectly by us). For example, elected officials determine which social benefits matter, the way in which these social benefits are to be quantified (e.g. unemployment rate, job growth rate, quality of jobs created, poverty rate), and they determine the price tag associated with these social benefits (e.g. the capital gains tax that the federal or state government foregoes by incentivizing investments in these communities and any administrative costs).

The TCJA, for example, indicates that investors will be allowed to delay paying a capital gains tax on any investments that are moved into an opportunity zone fund, will have to pay the capital gains on a smaller proportion of that initial investment depending on the number of years that the investment is held, and will not have to pay any capital gains on the proceeds from the opportunity zone investment. This is aptly summarized in a CNN article: "Here's how it works. Someone who reinvests a capital gain worth $100 in an Opportunity Zone in 2019 gets a 15% step up in basis," which means she has to pay the federal capital gains tax on only $85 of that original income. At a tax rate of 23.8%, that comes to $20 - and she doesn't have to pay it for another 10 years. On top of that, if she holds the investment for at least 10 years, she pays no capital gains taxes on the proceeds from the Opportunity Zone investment."

Governors were allowed to nominate Census tracts to become opportunity zones so long as they met one of the following criteria: (1) poverty rate of at least 20 percent; (2) median family income of the tract is 80 percent or less of the median family income at the metropolitan or state level; (3) contiguous to a low-income tract and does not exceed 125 percent of the median family income of the neighboring tract. Any tracts selected based on criteria (3) were to make up only 5 percent of the total opportunity zones in a state. There are no restrictions on the types of investments that can be made within opportunity zones other than "sin businesses" that include "liquor stores, gambling facilities, golf courses, country clubs, tanning facilities, and massage parlors".

But the effectiveness of these policies and whether they even produce the social benefits that we as a society care about is unclear. The decisions are made by elected officials but the degree to which they have been informed by the empirical economic literature is circumspect. Even Jared Bernstein, economist behind this TCJA provision and former chief economist to another Democratic presidential contender, Vice President Joe Biden, wrote: "If OZs [opportunity zones] turn out to largely subsidize gentrification, if their funds just go to places where investments would have flowed even without the tax break, or if their benefits fail to reach struggling families and workers in the zones, they will be a failure."

It is therefore unsurprising that these tax incentives in the Trump tax bill have received significant backlash (see here, here, and here for examples) if these incentives place a hefty cost to the federal government in foregone capital gains taxes - these capital gains benefits accruing to the wealthy, i.e. the 9.2 percent of taxpayers that report realizing any long-term capital gains at all - while their social benefits are reduced to these big "ifs". The California Budget and Policy Center estimates the cost to the federal government as follows: "These lost revenues - mostly benefiting high-income investors - could instead help pay for other services that may have a greater impact on vulnerable communities in California and across the nation. The official cost estimate for the tax incentives is small relative to the total cost of the TCJA - $1.6 billion over 10 years in a package of nearly $2 trillion in tax cuts. However, the long-run costs could be much greater given that this estimate does not include revenue losses from the complete exclusion of gains on QOF investments held for 10 years, which fall outside the 10-year period for which budget estimates were made." These costs and distributional implications are difficult to ignore.

To inform this debate from an economist's perspective, I focus on the evaluation of an existing tax incentive program for investments in low-income communities. The New Markets Tax Credit (NMTC) has been ongoing since it was legislated in 2000. Like the TCJA it provides incentives for investing in low-income communities but it places more stringent requirements on the ways in which investments are made.

Freedman (2012)

The NMTC program is distinct from opportunity zones in the TCJA because it provides tax credits to investors who make equity investments in Community Development Entities (CDEs). According to the federal government, any "domestic corporation or partnership that is an intermediary vehicle for the provision of loans, investments, or financial counseling in Low-Income Communities (LICs)" can qualify to be a CDE. However, in order to qualify these organizations need to have a primary mission of serving LICs and must maintain accountability to the residents of the LICs that they serve. Qualified CDEs can then take the equity investments and make Qualified Low Income Community Investments (QLICs). These requirements are stated very clearly in this presentation.

The NMTC program places more stringent requirements on qualifying investments than does TCJA. Not only does it require the investments to be made through CDEs (organizations must qualify to become a CDE) but it requires that the investments are made by the CDE to qualified active low income community businesses (QALICBs). NMTC does not allow these funds to be invested in businesses that build or rehabilitate residential rental property. This is a notable omission from the TCJA opportunity zone provision given the role that real estate development - particularly for rental property - plays in gentrification. It is discussed in part in this article on gentrification in NYC.

Freedman (2012) uses quasi-experimental variation to study the effects of NMTC. Because Census tracts at or below 80 percent of median family income of the metropolitan or state median family income qualify for NMTC-subsidized investments and tracts above 80 percent (even if they are 81 percent of the median family income) do not, Freedman uses a regression discontinuity (RD) identification strategy. His RD strategy allows him to estimate the causal effect of NMTC on community economic outcomes (poverty rate, median home value, median household income, unemployment rate, and household turnover) by comparing Census tracts around the 80 percent cutoff. In other words, the RD identification strategy assumes that Census tracts immediately above and immediately below the 80 percent cutoff do not differ based on any observable or unobservable characteristics other than the fact that those below the cutoff are eligible to receive NMTC-subsidized investments.

Freedman finds that most estimates are statistically indistinguishable from zero. In certain specifications he finds that: from OLS estimates $1 million NMTC-subsidized investment is associated with (1) .01-.03 percent decrease in median home values; (2) .02 percent increase in median household income; from IV estimates $1 million investment is associated with (1) reduction in poverty rates by one percent off a base of 13 percent; (2) reduction in unemployment rate by .33 percent off a base of 6 percent; and (3) increase in household turnover rates by .75 percent off a base of 16 percent. These select estimates are significant but very modest. He states: "Indeed, the results suggest that to the extent that there are benefits associated with subsidizing investment in poor areas, those benefits are limited, and for many outcomes we cannot rule out that there is no effect at all." The issue with Freedman's analysis is that it is conducted at the Census tract-level indicating that the results may be due to a change in the composition of the Census tract rather than an improvement in the economic outcomes of the existing residents. This is hinted at by the statistically significant and positive effect of investment on household turnover (i.e. there is more migration in and out of the tract).

The questions remain: (1) how does the value for money provided by this program - in terms of poverty reduction or one of its other goals - compare to value for money of other dedicated social programs? (2) how much of these results are driven by changes in the composition of neighborhoods (i.e. gentrification) as opposed to economic improvements seen by existing residents? The latter question is best answered with panel data at the individual or household-level rather than the Census tract-level. The former data would allow researchers to follow the same individuals or households between the pre- and post-investment periods and track the in and outflows from these communities.

Freedman (2014)

Question (1) above is arguably more difficult to answer but Freedman (2014) has another more recent paper that provides a partial answer to whether improvements in economic outcomes due to NMTC investments accrue to existing residents of these communities. While he does not use individual or household-level data to study the changing composition of the residents of a given Census tract, he uses administrative data to study the changing proportion of residents who work outside the tract and non-residents who work inside the tract. Given that policymakers would prefer that social benefits of these investments accrue to the residents of these zones rather than non-residents, this addresses an important dimension to the puzzle.

Freedman states in this paper: "a common feature of these programs is that, while restricting where businesses may locate or invest in order to receive subsidies or tax breaks, they place few constraints on whom subsidized businesses must hire... Further, to the extent that any new jobs subsidized under these programs fall into the hands of residents of distant communities, the local economic benefits of these programs may be diluted and any imbalances between the locations of jobs and housing exacerbated." He notes that only 30 percent of the state enterprise zone programs that he reviewed included an incentive for participating businesses to hire residents of those zones. This is not an incentive included in the NMTC or TCJA either.

The identification strategy in this paper is the same as in Freedman (2012). While Freedman (2014) does not use individual or household-level data he has data from the OnTheMap database constructed and maintained by the Longitudinal Employer-Household Dynamics (LEHD) program at the Census Bureau that enables him to study the proportion of workers in a tract who live in the same tract. These proportions are further broken down by different earnings categories and industries. This disaggregated data allows the author to connect changes in the number of composition of jobs to change in the proportion of jobs held by residents of the Census tract. He finds a small and - only in some specifications statistically significant - impact of NMTC investment on overall workplace employment. He specifically finds a statistically significant increase in employment in goods-producing industries with no impact on trade, transportation, utilities, or services employment which he indicates is consistent with previous findings (Harger and Ross, 2014) suggesting that NMTC attracted firms in capital-intensive rather than labor-intensive industries. This is suggestive that more limited skills in low-income communities may be deterring labor-intensive industries from locating in these communities.

Taken in conjunction with the effect on resident employment - a small decline in low-wage resident employment significant at the 10 percent level - these findings suggest that any positive effect of NMTC investments on employment accrue to residents outside of the Census tract. To explore this finding, the author further finds that observed changes in the commuting times of LICs were driven by an increase in commuters who were traveling at least 20 or more miles. As he states in his paper: "Households in these neighborhoods turn out to be not only physically distant, but also socioeconomically distant from households in LICs that receive investment. Indeed... there is a marked gradient in income levels and poverty rates as one moves away from tracts that received NMTC-subsidized investment. The figure shows median family income and poverty rates averaged across treated tracts and tracts at distances up to 30 miles away from treated tracts in the RD sample."

Taken together, this indicates that any modest benefits in job growth resulting from NMTC investment accrued to non-residents of the tract who were less-likely to be low-income. It should be noted - even acknowledging the positives of an Amazon HQ2 deal in Queens - that concerns like this were made by opponents of this deal. As stated in this Tech Crunch article on the issue: "Amazon's promise of 25,000 jobs (high-paying jobs) may have reduced that number [NYC unemployment rate], but there's no guarantee that those jobs would be filled by New Yorkers or Queens residents more specifically - and every indication that they would have gone to Amazon employees coming from somewhere else."

Policy implications

These are only two papers on the subject (notably both by the same author and employing the same identification strategy). However, the results from these papers are consistent with broader findings from the literature in that the evidence is very mixed and much of it is insignificant. This is similarly stated in the Minnesota House report: "Considering all the studies using regression analysis, the economic effect of enterprise zones remains unclear. Most studies find no significant increase in employment, while a few do. Moreover, the prospect for success seems greatest in already economically viable areas, rather than traditional zone locations - areas with stagnant or declining economies."

Ultimately, evidence of positive effects is inconclusive and, more importantly, there is at least some evidence of negative effects that would warrant a better investigation into the impact of investments on gentrification and displacement. Individual and household-level panel data can allow for such an investigation because they can provide greater visibility into the trajectory of existing residents of these communities. While the sum spent on this initiative is small relative to the cost of the broader TCJA, it is a large sum to be spent on a set of policies that has had mixed empirical support - and importantly, some of it negative - for several decades. These negative effects may also be exacerbated by the allowance of investments in residential rental property in this bill. Given the mixed evidence, these programs should collect data and evaluate the effectiveness of their program on social benefits but this is perhaps one of the most notable concerns about TCJA opportunity zones: there are limited reporting requirements and guidelines to ensure that investments are socially impactful. A bill on these requirements has been introduced in the House and should be followed closely.

  1. Freedman, M. (2012). Teaching new markets old tricks: The effects of subsidized investment on low-income neighborhoods. Journal of Public Economics
  2. Freedman, M. (2014). Place-based programs and the geographic dispersion of employment. Regional Science and Urban Economics
  3. Harger, K., Ross, A. (2014). Do capital tax incentives attract new businesses? Evidence from across industries from the New Markets Tax Credit. West Virginia University Working Paper. 
  4. Hirasuna, D., Michael, J. (2005). Enterprise Zones: A Review of the Economic Theory and Empirical Evidence. Policy Brief: Minnesota House of Representatives Research Department. 

Tuesday, June 25, 2019

Place-based economic policies (pt. 1 of 3): Motivation

My goal is to write a series of in-depth posts about geographic inequalities in developed countries and the evidence on the effectiveness of place-based economic policies. There were a few main motivating factors for this discussion: (1) increased attention has been paid to inequality within developed countries since the Great Recession and this discussion would be remiss if it did not address geography; (2) there is no doubt that economic geography has important implications for politics, particularly in the U.S.,where, for example, the Electoral College rather than popular vote governs the election of the president; and (3) place-based economic policies, including significant tax incentives for investments made in low-income communities introduced in the 2017 Republican tax bill, are gaining bipartisan popularity and should be evaluated as to whether they are effective at improving local labor market outcomes, health, education, and human development.

This has, however, proven to be a more difficult and time consuming endeavor than originally expected mainly because a conversation on this topic can cover a lot of material and go in many different directions (and is therefore difficult to organize). But - wanting to write something on this topic today on what would be chef, traveler, documentarian, and father Anthony Bourdain's 63rd birthday - this post will have to be the first in a series.

"Place" continues to be important even in our increasingly globalized world. As much as the world has become more and more connected there is no doubt that place continues to define - particularly for specific socioeconomic and demographic groups - what we can do for fun, what we can eat, where we can work, how much we can earn, and what our living standards are like. It continues to shape our experiences and how we view the world and our places in it. Industry specialization, for example, often defines the labor market opportunities and outcomes that are available in different regions.

I recommend this paper by Autor, Dorn, and Hanson (2015) that illustrates the impact of Chinese import competition on the U.S. labor market outcomes using a local labor market approach. The authors assign a measure of exposure to import competition from China to each local labor market area and utilize an instrumental variable approach to isolate the exogenous variation in this measure across local labor market areas. The reason that there is any variation in exposure to import competition at all is that there are differences in regional industry specialization patterns. These differences have therefore given rise to geographic inequalities within the U.S. whereby regions with import-competing manufacturing were particularly impacted by the growth in trade. These regions are illustrated in the below map taken from the paper. For a presentation on this paper and Chinese import competition you can see Autor's IFS Annual Lecture from 2017.

This is only one example of the way location continues to matter. Another particularly striking image comes from Raj Chetty's work on intergenerational mobility in the U.S. For more on this work, Chetty also presented at the IFS with Annual Lecture from 2014. The figure below illustrates the odds of reaching the top fifth of the income distribution for kids starting from the bottom fifth of the income distribution for metropolitan areas across the country. The differences by region cannot be ignored.

As stated in the New Oxford Handbook on Economic Geography, "This contribution [from economic geographers that economic processes have produced spatial differentiation and inequality] was crucial in counteracting hyper-globalist views, prominent in the 1990s, emphasizing homogenizing forces of globalization, envisaging a global society, and predicting the end of geography in economy, politics, and culture." Today would have been Bourdain's 63rd birthday and it is being celebrated in his memory as Bourdain Day. His show Parts Unknown illustrated the importance of "place" in an increasingly connected world.

Bourdain had a unique, respectful way of showing us - his global audiences - the local. He acknowledged and reported on the economic, social, and political structures that shaped the places he visited. He exhibited an unaffected empathy for the people he met and the lives that they led and this allowed him to share their stories in a way that we - people sitting thousands of miles away - could relate to. He had an effortless way of communicating with people underneath the layers and putting himself on the same level. At the end of the day he showed his audience the way people live in different places and - like a good social scientist - he mused on the reasons why they lived the way they did and how that was changing. Importantly, he wasn't afraid to call out injustices and criticize the perpetrators. A few of my favorite episodes in Parts Unknown - in Pittsburgh and West Virginia - address structural transformation and regional economies.

The decline in U.S. manufacturing - one consequence of globalization and to a lesser extent automation on the labor market - is only one of the components of ongoing structural transformation in the U.S. and other developed countries. Another important and related component is the relative decline in wages among those without a high school or college degree. The geographical significance of these changes is stated aptly in the Handbook, "Over the last eighty years, regional per capita income as a percentage of the national average showed signs of converging until the late 1970s. As much as anything, starting in the late 1970s, repeated recessions, major industrial restructuring and both age- and employment-related migration brought an end to the trend of convergence. Incomes and wealth began to concentrate in selected locations while bleeding out of others, reasserting the importance of places of economic power."

Yet, evidence of low labor mobility in the U.S. makes it more difficult for the population to adjust to changes brought on by structural transformation. Low mobility is hypothesized to to be a function of labor market institutions but it may also be due to changing ties to location and community. The ideas of location and community and what they mean are frequently raised in Bourdain's episodes, particularly in the context of immigrant communities and the decisions that people make to move across cities, states, and countries and their subsequent identities as immigrants. One of my favorite episodes followed Bourdain and fellow celebrity chef Marcus Samuelsson on a trip to Ethiopia. The episode captured the coexistence between Samuelsson's life as a global citizen - born in Ethiopia during the Civil War, adopted by a Swedish family at a young age, and immigrated to the U.S. to apprentice in a New York City restaurant in his early 20s - and his desire to connect with the country of his heritage with which he had little experience as a child.

Ultimately, much of our lives are still defined by the places that we are born and - whether we have the ability to choose or not - live. This is perhaps an unintended consequence of travel shows but was illustrated in Parts Unknown with unique attention to the "how" and "why". There is much that policymakers can do - and are trying to do - to reduce geographic inequalities where they exist. In many cases these efforts reshape the landscape of a place. These changes were discussed, for example, when Bourdain talked with Pittsburgh locals about local development initiatives and the changing identity of the city as it grows into a tech hub. How do these initiatives impact firm growth and investment decisions? To what extent do they lead to changes in labor market, health, and educational outcomes for existing residents? How do they impact labor mobility both in and out of the area? These are a few of the broad questions for future posts.

Wednesday, May 29, 2019

Jobs in developed economies

The Economist recently published an opinion piece on the status of work and jobs in developed economies that caught my attention. The topic of the piece - the success of today's labor market - is important due to its increased politicization and its implications for democracies in developed countries. Therefore this brief post will fact-check a few of the statements made in the article and provide a more comprehensive and accurate perspective on the labor market in developed economies. First, how well does the data support each of the following statements that were made in the article:
  1. Very low unemployment rates in developed economies - TRUE BUT MISLEADING. 
    • It is true that many developed economies have very low unemployment rates today. For an interactive look at the unemployment rate in OECD countries over the past several decades see here. Below is a screenshot from the interactive chart. This data shows that for many OECD countries, the unemployment rate today is lower today than at any point in the past thirty years. However, France and Southern European countries including Greece, Spain, and Italy are notable exceptions.
    • However, an analysis of the labor market that focuses on the unemployment rate without considering the labor force participation rate can be very misleading. I.e. is the unemployment rate very low because a significant proportion of the population has taken itself out of the labor market? Would these people accept a job if they were given one, despite not actively looking? In the U.S., labor force participation is lower than it was thirty years ago and it took a particularly significant hit in the aftermath of the Great Recession. In 1987, the labor force participation rate was 75 percent and it was 73 percent in 2017. The inactivity rate among working-aged men has risen from 14 to 20 percent over the same time period. Below is a screenshot from the interactive chart for labor force participation. This article explains why a recent rise in labor force participation this past year reflects a composition effect - meaning fewer people are leaving the labor force - rather than bringing people out of labor force back into it.          
    • The unemployment rate today is low by historical standards in many OECD countries - again the exceptions should be pointed out and the diversity within these developed economies should be acknowledged - but as an indicator it must be used in conjunction with the employment and labor force participation rates rather than as a standalone metric. 
  2. "Ever more women work" and women account for almost all the growth in the rich-world employment rate since 2007 - FALSE.
    • In the U.S., a decline in women's labor force participation is in part driving the overall decline mentioned above. This statistic adds a distributional dimension to this analysis that the article overlooks. In the U.S., specifically, labor force participation among women is lower today than in 1999 (which was the high point for women's labor force participation in the U.S.). For more on the declining labor force participation rate in the U.S., see this blog post on the topic. The below graph is taken from a BLS interactive page on labor force data. 
  3. "As for precariousness, in America traditional full-time jobs made up the same proportion of employment in 2017 as they did in 2005." - FALSE.
    • According to the OECD, full-time employment was 79 percent for men and 59 percent for women in 2005. The same figures for 2017 were 76 percent and 59 percent. 
    • Furthermore, just as unemployment rate was only one part of the picture to understand the trends in employment status, the full-time employment rate is only one part of the picture to understand labor market insecurity and the degree of job formalization. While this is not a statistic, this New York Times article from a few years ago exemplifies this point. It discusses the situation of two janitors: one who worked at Kodak in the 1980s and became a CTO of the company and the other who works at Apple today through a contracting company. The article details the aspects of job quality that are not captured by a simple full-time vs. part-time metric. In fact, OECD and ILO among other organizations have worked on measuring job quality. See Table 1 in this document that details the some of the measures of job quality: lifelong learning and career development, safety, ethics, working conditions, collective interest representation, and the stability and security of work. 
Not only do these statements provide a narrow and therefore misleading perspective on the labor market, building a case for a labor market that works for everyone requires research and evidence. The larger problem with this piece is that it does not put in the research and evidence required. Rather, it claims that now that the unemployment issue has been "settled" in developed countries (a questionable conclusion in and of itself), the public has moved on to a "series of complaints about the quality and direction of work" which are "less tangible and harder to judge than employment statistics." Most of our jobs as economists are to study things that are "less tangible and harder to judge".

George Akerlof has a forthcoming article in the Journal of Economic Literature entitled "Sins of Omission and the Practice of Economics" that discusses this bias against "important topics and problems when they are difficult to approach in a 'hard' way" with "hard" being defined as the ease or difficulty of producing precise work on a given topic. He argues that this bias within the discipline often leads to important topics being neglected at the expense of topics that can be more precisely studied. He provides several examples of situations in which the neglect of those topics that are more difficult to study precisely has led to an oversight in understanding. For example, he writes that in the lead-up to the Great Recession there were incentives to study the individual pieces of the recession puzzle but not to study them jointly.

On the theory side: "Following Caballero (2010), regarding theory, a model with all the pieces could not have been published; it would have been considered too far from precise, simple ideas (such as those that motivate simple new Keynesian or DSGE models); and, in this way, too Soft to merit publication." On the empirical side: "Regarding predictions from empirical evidence, the crucial data would have been of the wrong form... Even if she had uncovered, for instance, AIG's 533 billion dollars of commitments to insure securities such as CDS's, she would have still needed to turn it into the basis for a publishable paper. Those 533 billion dollars indicated tail risk of sufficient size to threaten a gigantic crash of the financial system; but it was only a single number. It was not the statistical evidence that typically underlies empirical papers in economics."

Akerlof's discussion has interesting parallels to this conversation on jobs in developed economies. Unemployment statistics are telling but they only tell one part of the story. The other part of the story - job quality, wages, and the working-poor rate - will likely better explain the dissatisfaction within developed economies that is driving the political shifts that we've seen in recent years. At the same time this other part of the story may be more complicated to study than unemployment statistics and, importantly for politicians, it may also be more complicated to discuss in the political arena. However, as the Times article poignantly illustrates, it should not be assumed that job quality has improved over the past few decades and the hypothesis that it has deteriorated - particularly for low-wage jobs - is a valid one that should be given its due consideration.

Tuesday, May 21, 2019

An all-in-one post for the past three months

Instead of doing a deep dive into one topic today, I have a few different points of discussion. First, thank you to Intelligent Economist for including me again this year in the top economics blog list. Second, I'll be joining a PhD program in Economics this fall and I can share my thoughts on the application procedure and offer whatever limited advice I have and hope/encouragement to those thinking about applying. This is particularly for those who have been out of school for more than a few years in job/grad school and those who found economics a little later in life (both of these apply to me). If I had one general piece of advice about PhD preparation, it is that I've found many people shy away from math and believe that only a few "select" individuals with innate abilities can be good at it (if I had a dollar for every economist I ran into while solving problems in a coffee shop who told me about the one genius in their college real analysis class) but - like anything else in life - I think those who are driven, purposeful, and work hard at it are well-rewarded.

One of the previous posts on this blog had discussed minimum wage policy. There wasn't enough time to cover all of the implications of minimum wage in that post, but I recently came across an interesting implication that I had not read about before. Specifically, a paper by Dettling and Hsu (2018) finds that higher minimum wages have significant effects on consumer credit markets (supply of unsecured credit, payday lending, and delinquency on credit payments). Higher minimum wages lead to lower borrowing costs for low income borrowers because they increase the number and favorability of credit card offers and they increase credit limits and decrease delinquencies. As noted in the paper, "labor market outcomes... are just one part of a household's finances. Interactions with consumer credit markets also play a crucial role in many families' economic wellbeing..."

Ethiopia gender diagnostic

The World Bank's Gender Innovation Lab - the team that I work for within the Office of the Chief Economist for Africa - has published a gender diagnostic report for Ethiopia. In this section, these views and interpretations are my own not that of the WB. The report does a few things: it provides evidence of gender gaps in agriculture, self-employment, and wage sectors in Ethiopia based on the Ethiopia Socioeconomic Surveys; it utilizes an Oaxaca-Blinder decomposition method to connect these gaps to gender gaps in the levels and returns to resources (e.g. fertilizer); and it provides concrete ideas to address the challenges that Ethiopian women face in the labor market. Oaxaca-Blinder decomposition decomposes the gender gap into observable differences in factors of production (endowment effect) and unexplained differences in returns to the same observed factors of production (structural effect). It allows us to determine to what extent differences in productivity are due to differences in the levels of resources versus the impact of those resources on productivity. It should be noted that the report is policy-oriented rather than academic in nature.

One example of a finding from this report is that the evidenced gender gap in agricultural productivity in Ethiopia is by and large due to unequal levels of productive factors such as land size and quality, fertilizer and other production inputs, formal credit, and farmer extension services (which can serve as a proxy for agricultural knowledge). When these - and other individual- and household-level observable characteristics - are controlled for, the gender gap in agricultural productivity drops from 36 percent to 6 percent. This is not necessarily the case in other countries in Sub-Saharan Africa, where giving female farmers access to the same level of productive factors as male farmers will not close the gender gap. For Ethiopia, we can assess how to close the gaps in factors of production.

For example, access to formal credit is an issue for not only female farmers but male farmers as well. The report on myths in African agriculture that I cited in an earlier post indicated that across the African continent only 6 percent of households used credit - formal or informal - to purchase agricultural inputs. It notes that "rural credit markets need to be deepened to serve farmers better, especially with respect to modern input use." On the other hand, the proliferation of farmer extension services is much greater with nearly 40 percent of male plot managers in Ethiopia having attended extension services recently (but only 23 percent of female plot managers having attended). There is a gender gap in both of these resources but while one has high take-up among male farmers, the other does not and therefore may require broader solutions.

If we focus on women's attendance of extension, we hypothesize based on existing literature and data that there are institutional factors that impact women's attendance and their level of agricultural knowledge more broadly. Namely, women are more time-constrained due to greater responsibilities in the home and are not as mobile due to costs of travel and to safety considerations. Both of these factors - time poverty and more limited mobility - can limit women's access to knowledge because they are not necessarily able to be in a particular place at a particular time to learn.

It is interesting because this underlying theme runs through many discussions about gender gaps in both the developed and developing worlds. A better understanding of how our existing systems are structured around the needs of specific subsets of our population can allow us to devise solutions that can better suit the needs of the others. For example, we posit that access to mobile phone technology can dramatically improve agricultural knowledge among female farmers because - conditional on their access to the technology - they will be able to access information at the time and place that is convenient for them (in Ethiopia, this is particularly challenging due to the limited competition in the telecommunications sector that has hindered mobile phone and internet penetration). Similarly, as this article in the Harvard Business Review illustrates, women in the developed world are advocating for more flexible working arrangements not to reduce hours but to manage workload at their own time and place where possible.

However, there are other important solutions as well. For example, investments into technologies that can alleviate the time poverty that women face in the first place. The tasks of collecting firewood, other fuel, and water for household energy consumption often fall on the women of the household and can take several hours per day in rural areas. Yet, there are interesting companies engaged in East Africa that are focused on addressing these energy consumption needs (some of which are highlighted in this report as examples). They provide alternatives to wood-fuel stoves in the form of solar energy or biodegradable biomass. This is just one example of how an evidence-based finding from the report can be developed to identify areas for future academic research, e.g. how successful are these alternative fuel companies and how effective are their alternatives at addressing women's time poverty? For more on these ideas, do check out the report.


There has been an upsurge in talk of recession recently in the popular media. It's not a topic that I've had much experience working on but I'll do my best to point out a few resources and start the conversation.

This article from the Fed lists the points of concern that have gotten analysts, investors, and economists talking in the first place. It lists four important housing market indicators, notes the significance of the housing market has in predicting economic downturns ("based on its forecasting track record - where a housing downturn is necessary but not sufficient for a recession to occur - the risk of broad-based economic recession certainly would be higher if the housing market were to weaken further"), and illustrates that recent trends are consistent with other pre-recessionary periods in 2001 and 2008.

The key is that current estimates of the four indicators listed - 30-year fixed mortgage rate, home sales rate, home-price change, and residential investment - are compared to their averages over the past three years to determine whether there is significant deviation from the average. For example, the below Fed chart shows the deviation in percentage points of the mortgage rate from the preceding three-year average. In the run up to the recessionary periods in 2001 and 2008 there was a rising deviation of the mortgage rate from the three-year average. The green trend for 2019 indicates the same pattern today.

Percentage-Point Deviation of 30-Year Mortgage Rate from 12-Quarter Average

It is unclear in my opinion that the other three indicators track the 2001 and 2008 trendlines as closely as they do here for the mortgage rate but either way I think this is one part of a larger picture. The larger picture is that in 2001 and 2008 these housing market indicators worked in conjunction with a private sector financial deficit (financial deficit from households and firms). This is discussed in an episode of the Exchanges at Goldman Sachs podcast - which I highly recommend - on five areas of credit market risk and how they impact the likelihood of recession. It is just one section from a GS report, "Learning from a Century of US Recessions."

The report itself provides a high-level summary of the various risk areas that lead to economic downturns but does not provide much detail on the individual risk areas. The podcast does a better job of discussing in detail the primary risk area: financial risk and asset bubbles. The GS viewpoint is that the current private sector surplus differentiates the current situation from 2001 and 2008 where the housing market may have been heating up - as it is today - but at the same time the private sector was running a large deficit. These two deficit periods are indicated in the GS figure below.

In the podcast, it is also mentioned that debt growth for households in the mortgage market is in decline - 16 percent inflation adjusted decline - that is unprecedented in the past 60 years. I ran a quick chart using the Fed's data to see the trendlines for all mortgage holders (in blue) and one- to four- family residences (in orange) and they do indicate a slow growth in recent years. How much slower than in the rest of the 30-year period shown here (1990-2018) is not clear since the trendline is still rising. However, it is unsurprising that mortgage debt is rising more slowly now than in the pre-2008 period given tighter credit standards in the aftermath of the Great Recession.

It is possible that an overheating of the housing market and relatively slow mortgage debt growth for households are consistent with one another if mortgage debt and home-ownership are more concentrated today than they were pre-2008. It would be interesting to see whether a smaller segment of the population is driving the uptick in the housing indicators being measured by comparing mortgage debt and home-ownership across the income distribution today and pre-2008. Tighter credit standards and more sluggish recovery among lower- and middle-income households after the 2008 economic downturn, in addition to rising income inequality in the aftermath of the downturn, may explain greater concentration in debt and home-ownership today. This could explain trends in the housing market as well as the private sector financial balance.   

Tuesday, February 5, 2019

Review of AEA sessions in Atlanta (Jan 4-5)

I took last month off from this blog (and most other productive activities) because I was on holiday for three weeks in the Bay Area. I hope all of you had a great holiday season 2018 with friends and family and a refreshing start to the new year 2019. The first topic I wanted to come back to is a review of the webcast sessions from the American Economic Association's annual meetings held in Atlanta from Jan 4-5, 2019. Several of the sessions are webcast here and you can access lectures on various topics including growth in the developing world, automation and the future of work, public debt, and - returning from last year with an extremely compelling panel - the gender problem in economics and what steps the profession can take to address it. In this post, I discuss two of the panels with an eye to discussing Autor's lecture on the future of work in the next post.

Growth challenges in the developing world 

The AEA convened a "World Bank economists" session consisting of three former World Bank Chief Economists (Justin Lin, Francois Bourguignon, Kaushik Basu), current Chief Economist Pinelopi Goldberg, and moderated by former Acting Chief Economist Shanta Devarajan. The purpose of the panel was to deliberate on the challenges facing the developing world. Given the very broad - arguably too broad - scope of the topic, it is natural that the panelists settled on a narrower topic over the course of the conversation: industrialization and the informality trap facing Africa.

Historically, industrialization and the rapid job creation in the formal wage sector that accompanies it have been seen as the most effective ways to raise wages and lower the poverty rate in developing countries. Lin cited historical examples of low-income countries' growth trajectories after capturing manufacturing jobs moving from the U.S. to Japan in the aftermath of WWII, Japan to Southeast Asia in 1960s and 1970s, and from Southeast Asia to China in 1980s and 1990s. Now that wages are increasing in China, many of these manufacturing jobs will be looking for a new home. How can Africa capitalize on these opportunities in coming decades was the question most of these economists were trying to answer. Chapter 2 of this policy report from the African Development Bank does a good job of summarizing these issues including evidence of what some economists call "de-industrialization" and the obstacles to small business growth. Given the demographic changes that will add 2 billion to the working age population in the African continent in this century, the creation of jobs in the formal wage sector will be important not only for economic but social and political stability. 
  1. The primary point of contention is that it is not clear that "de-industrialized" countries will capture these manufacturing opportunities without concerted policies. E.g. automation is a real threat to manufacturing jobs in certain industries and less so in others (retail incl. clothing, shoes, and furniture). Furthermore, the trade environment is rapidly changing with advanced economies looking to be less hospitable to imports from low-income countries. The second half of the panel asked panelists to comment on different ways of approaching this issue wherein I think the issue of too broad a topic came to light. I think it would have been more useful to showcase specific examples and evidence from recent research. 
  2. It wasn't discussed in the panel but it is relevant discuss the impact of a shift from self-employment and agriculture to industrial employment on working populations and whether there is desire on the part of working populations to hold these types of jobs in the first place. Specifically, J-PAL poses the issue in preface to a 2017 paper from Chris Blattman and Stefan Dercon that studied the effects of industrial employment on Ethiopian workers: "Industrial sector development to boost mass hiring is seen as important to poverty alleviation at the macroeconomic level. But how those jobs, particularly in early stages of industrial sector development, affect the workers themselves and what the workers prefer are less well-understood." The findings from this paper are summarized in this New York Times article with the bottom line being: workers are initially unaware but quickly become aware of the safety hazards and poor wages paid in sweatshop conditions leading to a high turnover rate in these early-stage manufacturing firms. The authors find that particularly when the constraints to self-employment were addressed through cash grants the workers preferred self-employment. 
      1. Does this mean that industrialization is not the best way to raise wages and lower the poverty rate in low-income countries? No. But it indicates that there may be a more efficient equilibria where a set of regulations providing a baseline level of safety for workers that address the issues identified in this study (chemical fumes, repetitive stress injuries, and probability of serious injury) can be beneficial to both employers via a lower turnover rate and to workers who would more likely work there if these health concerns were addressed. Such a set of regulations need not be so stringent that they reduce the comparative advantage of setting up shop in sub-Saharan Africa given the low wages on the continent but they will provide better standards of living for workers expected to drive these changes. 
Gender in the profession

On the panel on gender in the economics profession. The community by now is well aware of statistics indicating the low proportion of women who study economics as undergraduates, the lower proportion who study it as PhD candidates, and the even lower proportion who are tenured faculty at universities. The primary questions now, in my opinion, are (1) whether members of the community believe that these statistics are indicative of gender bias (as opposed to differences in ability or preference between the genders); and (2) whether members of the community believe that they can and should take action to address this bias, particularly when it is implicit and particularly where it requires the buy-in of economists who are neither part of the problem nor the solution.

Several of the questions posed in the panel revolve around these ideas. First is the need for data and evidence that is reflective of implicit bias to indicate to said economists that there is a problem at hand. Erin Hengel's paper on publication records of male and female economists that I discussed last year and Alice Wu's paper on sexism within the Econ Job Market Rumors website which is informal but commonly used among academic economists for job postings and career advice (see this interview with Wu on this paper) are two examples of this type of evidence. This webpage put together by the UC Berkeley Women in Economics group offers other useful information.

From my own anecdotes and research experience within the Gender Innovation Lab at the World Bank, there are a few issues that I think are actionable to address:
  1. Role models and social networks among women 
  2. Gender gap in perceived abilities in STEM fields 
  3. Culture and implicit bias within the profession
Given that the third issue is probably the one that is most difficult to address I think it requires first the buy-in from the community that I mentioned above. Being aware of implicit bias and its effects on the community are important because they are needed to take the next steps. For example, one issue that was talked about in the panel is aggression in economics seminars. It likely impacts women more than men because women tend to do better in collaborative and non-aggressive environments and the aggression tends to be more often directed towards women than it does towards other men (e.g. see Wu's paper on EJMR). But suffice it to say, I think we would all do better - men and women alike - if we were all a bit kinder to one another without compromising the rigor of our work. Specifically, to both acknowledge that we can and should be able to communicate questions and criticisms without resorting to aggression and be willing to learn the techniques to do so. Same with being willing to learn the techniques to recognize and address implicit bias.  

I have been supported in my efforts by peers and role model figures - mostly male - that have been enthusiastic about my ability to succeed in this profession. I have been blessed in not only role models in professional and academic life but also partners in my personal life that have been the most influential factors in my decision to undertake graduate studies. My thoughts on this issue are - in addition to addressing systematic issues within the field - if you can support a young person and believe in their abilities it is probably a determining factor in their decision to pursue higher studies. Whether we have the data or not as of yet (and there is more empirical research being conducted on role model figures and mentoring), we can't underestimate the value of empathy in how people decide whether or not they want to be in a particular location, field, university, firm. 

Thursday, December 6, 2018

In-depth look at income and wealth data (pt. 3 of 3): Minimum wage policy and the income distribution

In this final part of this series of posts on income and wealth, I originally intended to discuss the data used to analyze income inequality but I will introduce a more specific topic within income inequality: minimum wage policy and its impact on the wage and income distributions. Given the ongoing public debate over stagnating real wages despite a strong labor market (see this piece by the Pew Research Center for a concise description of the trends and a few of the reasons given by economists for the wage stagnation for workers at the lower end of the earnings distribution), it is particularly relevant to revisit the evidence on minimum wage policy and its impact on income inequality.

In this post, I focus on two papers - Autor, Manning, and Smith (2016) and Dube (2018) both in American Economic Journal: Applied Economics - that discuss the distributional implications of minimum wage policy. These papers have significant differences in both methodology and level of analysis. While AMS (2016) focus on individual wage inequality, Dube (2018) focuses on household income inequality with two iterations on how income is defined. The first is the conventional definition of income that includes both earnings and cash transfers. The second is a broader definition that also includes tax credits and non-cash transfers that enables the author to assess the substitutability between minimum wage earnings and government benefits to derive results that are closer to general equilibrium. The choice and level of the outcome variables measured in these two papers - earnings versus income and at the individual versus household level - are important to treat as distinct and independently informative. As I discussed in an earlier post on household income inequality in the U.S. and Britain, the distributions of individual labor market outcomes and household incomes do not necessarily track one another closely.


In the U.S., minimum wage policy is determined at several levels of government: federal, state, and as of late even citywide minimum wages exist (in San Francisco, San Jose, Albuquerque, Santa Fe, and Washington, DC). This piece on minimum wage policy outlines the basics. In part because the U.S. federal minimum wage declined in real value almost continuously for thirty years between 1979 and 2007 (it was fixed in nominal terms between 1981-1990 and 1997-2007), more than thirty states enacted legislation over the same time period to raise their state minimum wages above the federally mandated level. The below graph tracks the real value of the federal minimum wage (in part indicating the motivation for state and city-level legislation on the issue):

Source: UC Davis Center for Poverty Research (2018)
The minimum wage has therefore seen significant state and time-based variation within the U.S. over the past several decades that continues to this day. This variation has been utilized by many empirical studies of the minimum wage including AMS (2016) and Dube (2018).  It should be noted that the ratio of real minimum wage to median wage provides some quick information about lower-tail inequality that will be discussed in greater detail below. For example, between 1950-1970 this ratio fluctuated between 45-55 percent but by 1989 it had fallen to 36 percent.

Estimation strategy

The challenge with estimates of the impact of minimum wage policy are that both minimum wage legislation and wage inequality levels are impacted by several other unobserved characteristics. For more details on these challenges and how research designs can overcome them, see Allegretto et al. (2013). These two papers take two very different approaches: while AMS (2016) employ an instrumental variables strategy, Dube (2018) employs a series of sensitivities and falsification tests to indicate the robustness of his results. The takeaway is that identification of the employment and inequality effects of the minimum wage is challenging and can result in complicated empirical strategies.

As Dube (2018) indicates of the initial fixed effects model that he presents: "A problem with the two-way fixed effects model [state and time fixed effects] is that there are many potential time varying confounders when it comes to the distribution of family incomes. As shown in Allegretto et al. (2013), high- versus low-minimum wage states over this period are highly spatially clustered, and tend to be differ in terms of growth in income inequality and job polarization, and the severity of business cycles." In other words, the legislation of minimum wage and wage inequality are both impacted by a number of factors that are not controlled for in an OLS or even two-way fixed effects model.

AMS (2016) also reference potential biases when they present their initial OLS model. They cite confounding evidence that the effective minimum wage is found to be equally significant on both the lower-tail and upper-tail inequality (where it is expected to only have a significant effect on lower-tail inequality since minimum wage policy is only binding for at most the 15th/20th percentile of the wage distribution). The initial OLS model that they present estimates the impact of the "bindingness" of the minimum wage at the state-year level (a variable initially employed in Lee (1999)) - the log difference between the effective minimum wage and the median wage - on the difference between the log real wage at a specific percentile and the log real wage at the median. The former variable on the "bindingness" of the minimum wage is included as a quadratic term because minimum wage is expected to have a larger effect on the part of the wage distribution where it is more binding (i.e. at the lower-tail of the wage distribution) rather than a linear effect. To address potential biases, they employ an instrumental variable strategy and instrument for the observed effective minimum wage.

Because they include the effective minimum wage as a non-linear term, AMS (2016) utilize a set of three instruments as opposed to just the first one: (1) log of the real statutory minimum wage; (2) square of the log of the real minimum wage; and (3) interaction between log minimum wage and average log median real wage for the state across all periods. While it is not discussed in great detail in the paper, this instrument (1) I would assume is the legislated federal minimum wage and it clearly impacts the state-level effective minimum wage (either the federal or state minimum whichever is higher) but does not impact wage inequality at the state-year level through any channel other than the state-level effective minimum wage.

Dube (2018) does not employ an IV strategy in his paper on household income distribution but he includes several sensitivities and falsification tests of his original model to indicate the robustness of his results. His original model is the two-way fixed effects model that he critiques in the section above. In his "most saturated" specification he includes in addition to his original controls, division-specific year effects (to capture the effects of regional shocks on minimum wage legislation that may be driving some of the spatial heterogeneity that we see in minimum wage levels), state-specific recession-year dummies (to address the concern that minimum wage legislation is correlated with state business cycle fluctuations), and state-specific linear trends (to capture long-run trend differences across states). It is challenging to assess the effectiveness of these sensitivities and tests at obtaining causal estimates in comparison to quasi-experimental methods (seminal example is Card and Krueger (1993) in their study of the employment effects of minimum wage).


Based on their instrumental variables empirical strategy, AMS (2016) find that a 10 log points increase in the effective minimum wage leads to a reduction in the 50/10 inequality (inequality between the 50th percentile and 10th percentile in the wage distribution) by 2 log points for women, 0.5 log points for men, and 1.5 log points for the pooled sample. For a better understanding of why this paper uses log points as opposed to percentages, see this post on the topic. Women see a larger effect because a greater share of women work at minimum wage (6 percent of women in 2012 compared to 3 percent of men).

Source: Bureau of Labor Statistics (2013)
They are also able to state that the decline in the real minimum wage explains less than 50 percent of the rise in 50/10 inequality between 1979 and 2012 indicating that the majority of the rise in inequality over this period is due to changes in underlying wage structure (contradicting earlier findings that the decline in the real minimum wage contributed to around 60 percent of the rise in wage inequality over this period). This study indicates that erosion of the real minimum wage played a significant role in the growth of wage inequality over the past several decades but other factors, including skill-biased technological change and increased import competition from low-income countries which compose the "changes in underlying wage structure" that AMS (2016) reference, played a larger role than the decline in the real minimum wage.

What then of the related outcome of household income inequality? Dube (2018) finds that the poverty rate elasticity with respect to the minimum wage is between -.220 and -.552 indicating that a ten percent increase in the minimum wage yields between a 2.2 and 5.5 percent decrease in the poverty rate. He finds a first order effect of the minimum wage on the family income distribution of a ten percent increase in the minimum wage yielding between 1.5 and 4.9 percent increase in the pretax cash incomes of the 10th and 15th quantiles. This finding is indicated in the graphic below with the green line for cash income. The more interesting finding in my opinion is how much these minimum wage gains are offset by reductions in other government benefits due to ineligibility based on the higher income. He finds that for the bottom fifth of the income distribution, 30 percent of the gains to income resulting from minimum wage are offset by reductions in non-cash transfers and tax credits indicating that the relationship between wage and income (in its broader definition) is not nearly 1:1 for the poorest Americans. This is indicated in the graphic below with the orange line for cash income, tax credits, and non-cash transfers.

Source: Washington Center for Equitable Growth (2017)
  • These two papers' findings are consistent with one another though they measure different outcomes - individual wage inequality versus household income inequality. They both indicate that minimum wage has a significant impact on inequality (as well as absolute indicators such as the share of Americans living below the poverty line measured in Dube). 
  • It should be noted that the policy debate over minimum wage consists not only of discussion on the benefits of minimum wage policy on inequality, poverty, and other outcomes of interest but the costs in terms of the employment effects (i.e. hiring and firing decisions of firms, indicators of the quality of work, etc...). Economists are far from agreement on the employment effects of minimum wage (not discussed in this post). The extent to which minimum wage gains are offset by public assistance is also an important consideration and Dube's contributions to this question form part of a broader literature on the linkages between minimum wage and the existing social safety net. 
  • An important note about both of these papers is that by construction they reflect not only the mechanical effect of minimum wage (in increasing wages for individuals working at or below minimum wage) but may also reflect spillover effects on those who already work above it. This is an area for future research, however, since AMS (2016) attribute these additional effects not as spillovers but as measurement error of wages for low-wage workers. Specifically, they find that these spillovers do not "represent a true wage effect for workers initially earning above the minimum" rather accepting the null hypothesis that "all of the apparent effect of the minimum wage on percentiles above the minimum is the consequence of measurement error."
  • Increased trade and automation have yielded dramatic changes to the underlying wage structure of the economy. These changes may not be reflected in the share of workers who work at federal minimum wage (this share declined prior to 2000 and has remained roughly constant between 2000 and 2018 excepting a significant increase at the recession) but perhaps it has increased the share of those in the lower percentiles of the wage distribution. 
    • While these jobs in the lower percentiles may not be "minimum wage" jobs they are certainly part of a broader pattern of low-paying jobs in the U.S. These jobs - and the people who hold them - are illustrated in this article in the New York Times. The article indicates that nearly one-third of workers in the U.S. earn at or below $12/hour with 7.6 million Americans designated as "working poor" meaning they spent at least half of the year in question working or searching for work and were below the poverty line. 
  • Minimum wage policy will not - in a mechanical sense - impact these other low-paying jobs nor will it address factors that impact the underlying wage structure of the economy. However, in an environment in which trade and automation have changed the wage structure dramatically, it has been evidenced to address both wage and household income inequality. 
  1. Autor, D., Manning, A., Smith, C.L. (2016). The Contribution of the Minimum Wage to US Wage Inequality over Three Decades: A Reassessment. American Economic Journal: Applied Economics. 
  2. Dube, A. (2018). Minimum wages and the distribution of family incomes in the United States. Forthcoming in American Economic Journal: Applied Economics
  3. Allegretto, S., Dube, A., Reich, M., Zipperer, B. (2013). Credible Research Designs for Minimum Wage Studies. IRLE Working Paper #148-13. 
  4. Card, D., Krueger, A. (1993). Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania. American Economic Review

Friday, October 5, 2018

In-depth look at income and wealth data (pt. 2.5 of 3): A small note on wealth inequality from the archaeologist's perspective

I've been working on fellowship applications these past few weeks so naturally I began my research in a germane area of literature and ended up somewhere completely random. And by completely random I mean not even within the field of economics anymore and at best tangential to my original topic of investigation, but fascinating. I stumbled upon Ten Thousand Years of Inequality: The Archaeology of Wealth Differences, a volume on the archaeological studies of wealth inequality, and given its relevance to the posts I've been writing on inequality I figured I would make a small note on how inequality is being measured for a society that lived nearly two thousand years ago.

I haven't written a formal post on the Gini coefficient but given that this article uses it extensively in the archaeological context I preface by stating a few things: (1) the Gini coefficient is notably a simple measure of inequalities (most commonly income inequality) therefore it has its limitations that are well summarized in this Wikipedia post; (2) it has also been subject to revision and extensive debate as well as the creation of alternative measures of inequality including the Atkinson Index that may be more informative if certain contexts; (3) given my lack of knowledge of archaeology (and my lack of knowledge more broadly on the range of applications that the Gini coefficient has had in diverse fields within social science) I don't assess whether or how the Gini coefficient was applied and rather introduce it as a thought-provoking application outside of the realm of economics.

Feinman, Faulseit, and Nicholas (2018) provide estimates of wealth inequality for the Classic period in the history of the pre-Hispanic Valley of Oaxaca, Mexico based on archaeological house excavations at six pre-Hispanic settlements. They rely principally on architectural constructions and space to proxy for wealth and apply the Gini coefficient to three architectural variables: terrace area, house size, and patio area. They also utilize distribution of artifacts such as obsidian and other rare items.

The Lorenz curve in the figure below shows the Gini coefficient constructed based on the house sizes for all of the houses in the sample (a total of 36 excavated houses across all six sites in the Valley of Oaxaca) with a coefficient of 0.35 and a 95 percent confidence interval between 0.31 and 0.39. A Gini coefficient of 0 represents perfect equality whereas 1 represents perfect inequality.

The Gini coefficients from all samples (houses, patios, and terraces) and excavation sites are indicated in the figure below. They range from 0.35 to 0.43.

The authors find based on their analysis that wealth inequality during this time was low compared to other urbanized and preindustrial settings (confirming extant evidence).

While there was notable variation between the periods that the authors link to changes in the socio-political structures of the time, they specifically note that "[t]he consistently low Gini values... are informative, especially as indicators of wealth inequality, because they challenge the long-term notion that archaic states were always starkly divisible into the rulers and the ruled, with dramatic differences in resources and quality of life between the two. This coercive/despotic vantage on archaic states is well ensconced in the historical/social sciences for preindustrial times (e.g. Mann 1977; Wittfogel 1957) but now is being challenged as not uniformly applicable (Blanton 2016; Blanton and Fargher 2008), with some historical polities seen as having had a more collective institutional orientations and lower degrees of wealth inequity (e.g. Mann 2016, for a change from his earlier perspective)."

A few comments and questions came to mind about the application of Gini coefficient in this context:
  1. The sets of data used in these analyses of the Classic period in particular are considered by the authors to be large and representative (likely given the difficulties involved in excavations and the number of houses, patios, and terraces that are still intact after thousands of years) but they are still subject to the well-described small sample size bias associated with the Gini coefficient. Smaller samples are biased towards having smaller Gini coefficients which may make it difficult to compare excavated sites in the graphic below (from the Smithsonian Magazine article about this book) with the United States today, which is not based on excavated evidence and has a much, much larger sample size. 
  2. Comparisons across past civilizations, though, if they are based on similar sample sizes may be more informative than comparisons between past civilizations that were excavated and modern day societies. Similarly, I would find variation within a given civilization over time (as evidenced in the article) to be informative especially in conjunction with changes in socio-political structures. This is hinted at by the authors when they quote from Piketty (2015) on the impacts of these structures on inequality: "[O]ne should be wary of any economic determinism in regard to inequalities of wealth and income... The history of the distribution of wealth has always been deeply political... How this history plays out depends on how societies view inequalities and what kinds of policies and institutions they adopt." 

Wednesday, October 3, 2018

In-depth look at income and wealth data (pt. 2 of 3): Wealth

First, to preface with why wealth as distinct from income is relevant to economists and to policymakers at large. Kopczuk (2014) discusses the importance of understanding the wealth distribution: "the extent to which the well-off are going to rely on work vs. return to their wealth in the future is clearly important for assessing the extent to which a society will view itself in some way a meritocracy." Wealth is an important determinant of labor force participation and therefore impacts productivity and economic growth. It also has important implications for inequality, intergenerational mobility, and, consequently, implications for democratic institutions whose stability is reliant on a meritocratic society or at least the verisimilitude of a meritocratic society.

It should be noted that estimates of wealth inequality and the top wealth shares are not as widely agreed upon as estimates of income inequality and labor income shares. There are a few main data sources for estimating wealth inequality that are aptly summarized in Alvaredo, Atkinson, and Morelli (2018):
  1. Household surveys including the U.K. Wealth and Assets Survey and the U.S. Survey of Consumer Finances;
  2. Administrative data on individual estates at death; 
  3. Administrative data on wealth of living from annual wealth taxes; 
  4. Administrative data on investment income that are capitalized; and 
  5. Lists of large wealth-holders (e.g. Forbes).
These data sources are discussed in great detail in Kopczuk (2014)'s "What Do We Know About the Evolution of Top Wealth Shares in the United States?" which specifically discusses the U.S. Survey of Consumer Finances (1), the mortality multiplier method with individual estate data (2), and investment income data (4). Each of these data sources is subject to different concerns. Household surveys and list of the wealthiest individuals are recent phenomena and cannot be used for estimates prior to the 1950s when the household surveys on wealth were first implemented. Administrative data on wealth of the living based on wealth taxes cannot be recouped in most developed countries because only a few developed countries, most notably France and Norway, have a wealth tax to begin with. Therefore, most researchers rely on estate tax records on individual estates at death or on reported taxable capital income.

The primary concern with estate taxes is that the distribution of estates of the deceased must be projected to the population at large: i.e. a multiplier method must be used in order to answer the question, how does the distribution of wealth among the deceased reflect the distribution of wealth among the living? Mortality multipliers are inverses of mortality rates based on various criteria, for example, wealthy individuals tend to have lower mortality rates and increased longevity compared to less wealthy individuals and therefore a higher mortality multiplier would be applied to the upper estate ranges meaning there are relatively more individuals living within those ranges than lower ones. For more on recent discussions of the relative longevity of the wealthy see Saez and Zucman (2016) and Chetty et al. (2016).

Kopczuk presents a few interesting stylized facts about wealth that provide a good introduction to the wealth distribution and methods of estimating it:
  • Wealth is highly concentrated (top 10 percent holds between 65 and 85 percent of the total wealth, top 1 percent holds between 20 and 45 percent of total wealth based on time period); 
  • While the methods of estimating the wealth distribution disagree on the timing it is clear that wealth concentration hit its apex prior to the Great Depression and declined after that; 
  • Different methods lead to varying estimates for the top 1% for several reasons: one is that the estate tax multiplier method uses the individual as the unit of observation, surveys use the household, and the capitalization method uses tax units; another is that tax evasion impacts the administrative tax-based methods (estate tax and capitalization) but not the survey-based methods. Some capture debt (estate tax returns) whereas others do not (capitalization). 
In a recent issue of the Journal of Public Economics commemorating Tony Atkinson's work, Alvaredo, Atkinson, and Morelli (2018) provides new evidence on the evolution of top wealth shares in the U.K. To choose one of the most interesting facets of the discussion of wealth that they present in the article, it is enlightening to view the top wealth shares compared to the wealth shares excluding housing.

The top 1%'s total wealth share and wealth share excluding housing tracked each other for much of the late 20th century but the authors note the divergence between the two trends in the 21st century, wherein the share of the top 1% of wealth holders of total wealth increased much more rapidly than its share of wealth excluding housing. In other words, the growth of wealth excluding housing is likely to be a more significant contributor to rising inequality than is the growth of housing wealth. In fact, they even mention that increases in housing prices serve an equalizing effect for the top 1%:

"It appears that housing wealth has moderated a definite tendency for there to be a rise in recent years in top shares in total wealth apart from housing. When people talk about rising wealth concentration in the U.K., then it is probably the latter that they have in mind... The results show how the impact of a general rise in house prices has changed over the period but it is always equalizing for the top 1%. At the beginning of the period a rise of 25% led to a reduction of some 1 percentage point in the share of the top 1% but the effect became smaller over time."

It should be noted, however, that trends in the housing market - particularly the resurgence in the private landlord and "buy to let" over the past three decades - likely have impacts on other areas of the wealth distribution apart from the top 1% of wealth owners (though these impacts are not addressed in this paper). This New York Times article from last year, for example, is a news feature that discusses the role that homeownership plays in propagating existing wealth and income inequalities. These topics and the lower rungs of the wealth distribution more broadly are areas for further investigation, but for the time being Alvaredo, Atkinson, and Morelli (2018) highlight how granularity in wealth data can be used to better identify the causes of growing wealth inequality over the past few decades and, while they utilize estate data and the mortality multiplier method in their analysis, can also be triangulated with other methods and data sources to form a more comprehensive understanding of the wealth distribution.

  1. Alvaredo, F., Atkinson, A., Morelli, S. (2018). Top wealth shares in the UK over more than a century. Journal of Public Economics.
  2. Kopczuk, W. (2014). What do we know about the evolution of top wealth shares in the United States? NBER Working Paper 20734.
  3. Chetty, R., Stepner, M., Abraham, S., Lin, S., Scuderi, B., Turner, N., Bergeron, A., Cutler, D. (2016) The association between income and life expectancy in the United States 2001-2014. Journal of American Medical Association.
  4. Saez, E., Zucman, G. (2016) The distribution of US wealth, capital income, and returns since 1913. Quarterly Journal of Economics