1. Skip to navigation
  2. Skip to content
  3. Skip to sidebar

Economic Currents

Keep up to date with the latest UHERO news.

UHERO Fellow Interview Series: Tim Halliday

Posted September 4, 2014 | Categories: Q&A, Blog

Sumner La Croix interviewed UHERO Fellow Tim Halliday about his Social Science and Medicine paper in July 2014. For more on this paper, see Tim's blog post here.

1. Tell us something about yourself ...

I earned my PhD from Princeton in 2004. I have been at UH-Mānoa since then. I am also a fellow at the Institute for the Study of Labor Economics (IZA) in Bonn, Germany. My research lies in the field of human resource economics, which encompasses labor, population and health economics and tends to be very data intensive. These days I have been working a lot on inequality in various guises. One recent project is on the evolution of wage distributions in the United States and Mexico since the latter part of the 1980's. Another uses Bayesian econometric techniques to estimate the inter-generational transmission of health status.

 2. Is this a good summary of your results: Unemployment kills!

More-or-less. This work seems to be suggesting that poor macroeconomic conditions do increase mortality risks but only for working-aged men. There is no such association for the elderly or for women. In some way, this makes sense since working-age men have the strongest attachment to the labor force. One important point is that this is one of the few studies that uses individual level data; others typically use aggregate state-level mortality rates which can be hard to measure. These studies actually show the opposite, namely, that poor macroeconomic conditions are associated with lower mortality, even, for the elderly. While we do not understand why there is this difference between the results at the differing levels of aggregation, we can say two things for sure. First, mortality is very easy to measure at the individual level; you are either alive or dead and that is easy to verify. On the other hand, a mortality rate for a given state is actually hard to measure because it is defined as the number of deaths in that state during a given period of time divided by the states population at a point-in-time. The fact that the denominator is moving is what makes this a challenge. Second, related work has shown that job displacement kills you. This work also uses microdata. It is a lot easier to reconcile my findings with this literature than the "recessions are good for you" literature with it.

3. Were you surprised to find such a big effect?

Yes, but there really isn't another study out there that does exactly what I did, so there is not a comparison that we can make. I find that a one percentage point increase in the unemployment rate results in about 24 more deaths per 100,000 workers which in epidemiological terms is quite large. However, the US economy has business cycles which means that the unemployment rate goes up and then comes down. So, over a prolonged period, on net, my estimates would indicate a smaller number of deaths since some years would be good, but that it is so responsive was surprising. Bottom line is that more work needs to be done using other large individual level data sets from the US to see what we get in other contexts.

4. How did you get interested in this topic?

When I was starting out in graduate school, I had initially wanted to work in development and much of my work is in developing countries. However, at that time, many of my advisers who had been working on savings and consumption started to think about health and how it fits into life-cycle economic behavior. This actually seemed like a natural progression since health is probably the most important component of human welfare. So, during one meeting with my adviser, Chris Paxson, she had mentioned Chris Ruhm's work on recessions and health tangentially. This work is pretty much the consequence of that specific interaction.

5. Are there policy implications?

Yes. In fact, Harvey Brenner of Johns Hopkins who is probably the godfather of this literature has testified before the US Congress on numerous occasions. Although I am not sure if knowing that recessions increase mortality risks makes good stewardship of the macro-economy any more of a moral imperative; perhaps it does but it would be important even without these mortality effects. These results would possibly affect other cost-benefit calculations. For example, environmental regulations will confer long-term benefits down the pike but one immediate cost that might be ignored could be these mortality effects if the regulations increase unemployment.

Actually, I presented this paper once and Edward Lazear, who was the second George Bush's chief economic adviser, was in the audience. He told me that the auto bailout, of which he was the chief architect, probably prevented the unemployment rate from going up a full percentage point. He said that my work tells how many lives this policy saved.

How Do We Measure Social-Ecological Resilience?

Two UHERO graduate researchers, Alex Frost and Cheryl Scarton, attended a field course about social-ecological resilience of island systems in Nadave, Fiji. Participants of the field course were students and environmental practitioners from places throughout the Pacifc like Fiji, Vanuatu, Micronesia and the Solomon Islands.

On day three of the field course, the group took an early morning boat ride to Viwa, an island community of 30 households that is largely food and water self-sufficient. The home stay experience immersed participants in a traditional village lifestyle to apply terrestrial and marine survey methodologies learned from previous days.

Based on western standards and traditional economic measure like income, labor, and production, Viwa would be considered impoverished. The residents’ primary income is through selling excess fish and crops at the market and organization of a home stay immersion program. The island has intermittent power at night from a diesel generator, water comes from a thoughtfully engineered catchment system, and the intermittent power limits access to television and Internet.

From a lens that focuses on social and natural capital versus human and financial capital, Viwa is ecologically and socially wealthy. There is strong community cohesion - every Monday is a rotating social work day, where residents take the time to help one family plant, weed or harvest. Everybody shares excess harvest and people have time for leisure and storytelling. They are always joking, laughing and singing. The knowledge of agroforestry and management of fisheries is passed down through observation and application between generations that do not harm the health of the soil and encourages biodiversity, both are key indicators of sustainability.

How resilient is Viwa island? The common definition of resilience is “the capacity of a system to absorb disturbances or shocks and adapt accordingly while still retaining the same function and structure (McClanahan et al. 2012).” Economically speaking, global financial collapse will probably not affect Viwa at all, but the increasing demand of marine resources from Asia is pressuring people to over harvest. Ecologically, it seems the biggest challenge is invasive species, but heterogeneity of the agroforestry system minimizes the spread of pests and disease, compared to monoculture agriculture. Additionally, they have social mechanisms in place to prepare for extreme events like hurricanes. The island is especially vulnerable to sea level rise, coral bleaching, and shifts in weather patterns (such as a long drought). The village residents are working to develop an extensive water infrastructure system in the future to connect water pipes with an adjacent island. Still, the village faces many social challenges. Younger generations are adapting to the expectations of the market economy by working off island, which leads to a loss of traditional ecological knowledge. Concurrently the growing island population requires further clearing of the land for more housing.

 Overall, the week-long intensive field course brought together faculty, students, and experts to disseminate and learn various methods and tools to measure social-ecological resilience. The forum encouraged network building between the University of Hawai’i and University of South Pacific. The variety of perspectives helped increase participant capacity to challenge existing mental models and assumptions. The experience inspired students to develop future interdisciplinary research on island resilience and identify opportunities to mitigate complex challenges that face Pacific nations, like the impacts of climate change.

- Alex Frost and Kim Burnett

UHERO 101.12: What is the Value of the Environment?

The Earth’s environment is divided into different combinations of living organisms and their nonliving surroundings: air, water and soil. These different organic communities are called ecosystems. Humans receive benefits from these ecosystems in the form of “ecosystem services”, a term that covers a range of benefits from artistic inspiration to soil detoxification. (See below for a list of example ecosystem services)*

In 1997 Robert Costanza and 12 other authors wrote an eye-opening article in Nature called “The value of the world’s ecosystem services and natural capital” (Costanza et al. 1997). It stoked interest in environmental valuation because of the $46 trillion/year value (in 2007 US dollars) it placed on the planet’s services. After factoring land use change the value of Earth's services in 2011 was updated to $125 trillion/year (in 2007 US dollars) (Costanza et al. 2014). The goal of Costanza et al. (1997) was not to commodify the environment, but more so to raise awareness to what these environmental benefits are worth in a capitalist market economy. The methods for arriving at these dollar figures were questioned and the valuation was controversial because some people are naturally inclined to ask…

How can you put a dollar value on the environment?

Putting aside the ethical question of assigning dollar values to experiences and connectivity with other people and nature, the below table sums up how previous academic research has addressed environmental valuation: 


These valuation methods are further described in Module 4, Session 2 of this training resource from The Economics of Ecology & Biodiversity (TEEB).

Prices of goods and services sold in markets can be used to arrive at a dollar value for certain aspects of the environment, but in the case that market values are not available, non-market based methodologies have been used to arrive at a value. Ecosystem service valuation is a relatively new field and researchers are collecting results from previous studies to help future researchers confirm what valuation methods work best for different ecosystem services (Ecosystem Services Valuation Database). A paper was written by De Groot et al. (2002) which includes a table of ecosystem functions and their compatibility with different valuation techniques to help guide in assigning a dollar value to ecosystem services. With some ecosystem services there are intrinsic values (such as existence values) that are hard to put into dollar terms. It doesn’t always have to be about money....

There is more than one way to value the environment

Ecosystem services do not have to be valued in terms of dollars. Any unit can be the common denominator such as time, energy, or freshwater, for example. Environmental valuation differs from financial valuation in that it is rarely done to account for an entity’s profit, it is done to account for alterations humans have made on the environment, or to help decision makers evaluate consequences of their actions. Farber et al. (2002) defines valuation as an assessment of trade-offs toward achieving a goal such as reduced carbon emission, increased habitat or improved water quality.

An important concept to keep in mind is that people do not directly benefit from ecosystems without human, social and built capital. The valuation of the environment’s natural capital must be parsed out from the entire interaction between people, communities and their built environment. It is only through institutions as well as human management and invention that we extract benefit from nature (Costanza et al. 2014). The scope, precision, techniques and units used in an environmental valuation depend on the purpose. Ecosystem service valuations are done at different spatial scales to suit different objectives such as raising awareness, national income and well-being accounts, specific policy analyses, land use planning, payment for ecosystem services, full cost accounting and common asset trusts. For more information on the field of ecosystem service valuation check out references below:

Ecosystem Services:
The Ecosystem Services Partnership
The Economics of Ecosystems & Biodiversity Initiative

Referenced Papers:
Costanza, Robert, Rudolf de Groot, Paul Sutton, Sander van der Ploeg, Sharolyn J. Anderson,
          Ida Kubiszewski, Stephen Farber, and R. Kerry Turner. 2014. “Changes in the Global
          Value of Ecosystem Services.” Global Environmental Change 26 (May): 152–58.           doi:10.1016/j.gloenvcha.2014.04.002.

De Groot, Rudolf S., Matthew A. Wilson, and Roelof MJ Boumans. 2002. “A Typology for
          the Classification, Description and Valuation of Ecosystem Functions, Goods and           Services.” Ecological Economics 41 (3): 393–408.

Farber, Stephen C., Robert Costanza, and Matthew A. Wilson. 2002. “Economic and
          Ecological Concepts for Valuing Ecosystem Services.” Ecological Economics 41 (3):

Robert Costanza, Ralph D’arge, Rudolf de Groot, Stephen Farber, Monica Grasso, Bruce
          Hannon, Karin Limburg, Shahid Naeem, Rpbert V. O’Neill, Jose Paruelo Robert G.
          Raskin, Paul Sutton & Marjan van den Belt. 1997. “The Value of the World’s Ecosystem           Services and Natural Capital.” Nature 387 (May): 253 – 260.

 - Cheryl Geslani

*(List comes from the Ecosystem Services Valuation Database)

Air quality regulation Fish Pollination of crops
Animal genetic resources Flood prevention Prevention of extreme events [unspecified]
Artistic inspiration Fodder Provisioning values [unspecified]
Attractive landscapes Food [unspecified] Raw materials [unspecified]
Biochemicals Fuel wood and charcoal Recreation
Biodiversity protection Gas regulation Refugia for migratory and resident species
Biological control [unspecified] Genetic resources [unspecified] Regulating [unspecified]
Biomass fuels Hunting / fishing River discharge
Bioprospecting Hydro-electricity Sand, rock, gravel. Coral
C-sequestration Industrial water Science / research
Capturing fine dust Inspiration [unspecified] Seed dispersal
Climate regulation [unspecified] Irrigation water [unnatural] Soil detoxification
Cultural use Maintenance of soil structure Soil formation
Cultural values [unspecified] Meat Solar energy
Decorations / Handicrafts

Microclimate regulation

Spiritual / Religious use
Deposition of nutrients Natural irrigation Storm protection
Disease control NTFPs [food only!] TEV
Drainage Nursery service Timber
Drinking water Nutrient cycling Tourism

Dyes, oils, cosmetics (Natural raw
material for)

Other ESS Various
Ecotourism Other Raw Waste treatment [unspecified]
Education Pest control Water [unspecified]
Energy other Pets and captive animals Water Other
Erosion prevention Plants / vegetable food Water purification
Fibers Pollination [unspecified] Water regulation [unspecified]
Fire prevention    


How Do We Manage Our Interdependent Environmental Resources?

Managing water resources requires an understanding of the linkages between key hydrologic factors and direct human influences. The problem is further complicated by the fact that water resources are often interdependent, which suggests that management should also account for ecological interlinkages. For example, a forested upstream watershed may replenish an underlying groundwater aquifer, or a coastal groundwater aquifer may provide positive spillover effects to a downstream nearshore resource such as a fishery. Left unregulated, these spillover effects are economic externalities—additional, unintentional costs or benefits. In general when private parties act in their self-interest in the presence of externalities, the outcome may not be the best for society.

The Kukio Region: Groundwater and Limu

In an application to the Kukio region on the Big Island, Pongkijvorasin et al. (2010) explore how the relationship between submarine groundwater discharge (SGD) and a keystone algal species, Gracilaria coronopifolio (“limu”), in the nearshore affects optimal water management. Lab experiments suggest that moderate levels of SGD influx to a coastal marine environment increase the growth rate of limu due to resulting changes in nutrient loads, temperature and salinity (Duarte et al., 2010). A reduction in the aquifer, and hence SGD, generates a negative externality since there is less water entering the coastal environment. This study shows that optimal water management before accounting for the limu involves only slightly higher water pumping rates (roughly 6 million  per year over 100 years in both cases) because the market value of algae is relatively small compared to the benefits of water consumption. However, the market value of limu does not include ecological and cultural values. One way to account for values that are difficult to monetize is a minimum algae-level constraint. If the stock of limu is constrained to be no less than 90% of its current level, the effect on optimal extraction rates is much more dramatic: extraction starts at approximately 4 million  per year, falls to 3 million  annually by year ten, and stabilizes at less than 0.5 million  per year from year 22 onward.

Incentivizing Externalities

Once we understand how optimal resource extraction rates change in the presence of an externality, the next question is how do we internalize it? In other words, what can we do to incentivize private actors (e.g. water consumers) to behave in a way that provides the most benefits to society? When the externality is negative, as is the case where reducing the groundwater stock slows limu growth in the nearshore, a corrective tax can be implemented to reduce groundwater extraction and increase the benefit of higher groundwater levels over a longer period of time. When the externality is positive, as is the case when watershed conservation activities increase recharge for a downstream aquifer, the socially optimal level of conservation can be incentivized using payments or subsidies. As the number of positive and negative externalities within a water management system increases, so does the complexity of the optimal tax/subsidy formula. Nevertheless, advancing methods for managing linked natural systems is important, especially in the context of water resources, given trends of increasing scarcity worldwide and the expected effects of climate change.

-Christopher Wada



Duarte, T.K., Pongkijvorasin, S., Roumasset, J., Amato, D. and K. Burnett (2010), ‘Optimal management of a Hawaiian Coastal aquifer with nearshore marine ecological interactions’, Water Resources Research46, W11545.

Pongkijvorasin, S., J. Roumasset, T.K. Duarte and K. Burnett (2010), ‘Renewable resource management with stock externalities: Coastal aquifers and submarine groundwater discharge’,Resource and Energy Economics32, 277-291.


Income Inequality in Hawaii Since Statehood

Posted June 5, 2014 | Categories: Hawaii's Economy, Blog

Thomas Piketty’s best-selling tome on the evolution of inequality in the US, Capital in the Twenty-First Century, has inspired us to ask—how does the distribution of income in Hawaii compare with that in the country as whole?  And how has that distribution changed over time?  To answer these questions, we construct an extended time series of data on income distribution and measures of income inequality for Hawaii. Using tax data from the Internal Revenue Service (IRS) and the State of Hawaii Department of Taxation (HIDOT), we compare various measures of inequality in the US and Hawaii.*

Income Share by Income Class

Here we use the labels used by Piketty to describe various groups in the income distribution. The “Dominant Class” is made up of the top one percent of families; the “Well-To-Do Class” is the group of the nine percent of families below the “Dominant Class”. The next 40 percent is called the “Middle Class” and the remaining 50 percent the “Lower Class.”

Looking at the IRS and HIDOT data,**
 we can see the share of income going to the “Dominant Class” increases while the share going to the “Lower Class” decreases steadily from the mid 1950s up to the present for both the US and Hawaii.   However, we also see that the income distribution in recent years is more compressed in Hawaii than the US.

 focus in on each of these classes one at a time to more cleanly compare the pattern of inequality in Hawaii to that in the US as a whole.  

The share of income claimed by the top one percent has been lower for Hawaii than for the US overall since the early 1980s. Up until the mid 1990s the share of income claimed by the bottom 50 percent in Hawaii lagged behind the share claimed by the bottom 50 percent in the US overall.

The Gini coefficient is one measure of the distance between a perfectly equal distribution of income and the actual distribution of income. A Gini coefficient of zero represents a perfectly equal society while a coefficient of one occurs when one individual claims all the income and everyone else claims none. Since the mid 1950s, the Gini coefficient has steadily increased from 0.43 to around 0.60 for both Hawaii and the US as a whole.

Income Mobility

One important caveat in this analysis is the importance of mobility. If those who were in the “Lower Class” last year are in the “Well-To-Do Class” this year, inequality is merely temporary.  Unfortunately, recent research on inequality in the US and various Northern European countries suggests that greater inequality of incomes in the present leads to reduced intergenerational mobility in the future (Corak, 2013). 

The Equality of Opportunity Project
 hones in on this issue of income mobility and the New York Times produced an interactive map for exploring the data. From this we can see that a child raised in Honolulu whose family income was in the bottom 20% had a 10.1% chance of reaching the top 20% of family income in adulthood. 

Forces Behind Distributional Changes

One important finding in our analysis is that, since the mid-nineties, the share of income going to the top 1% is lower and the share going to the bottom 50% is higher in Hawaii than in the US.  The fact of the matter is, economists don’t fully understand all of the causes of inequality, and a full analysis of the causes for Hawaii is well beyond the scope of this blog post.  We can, however, speculate on potential causes based on the rapidly growing research on inequality.

First, higher levels of unionization in Hawaii may be protecting wages in the bottom of the distribution.  In addition, recent evidence from Autor, Dorn and Hanson (2013) has shown that China’s entry into the WTO has severely hit wages of low skilled workers in the US, so the absence of a large manufacturing sector in Hawaii may be responsible for the higher income shares in the bottom half of the distribution in Hawaii vis-à-vis the US mainland.

The top 1% is not as well understood.  Some such as Piketty have speculated the higher top income shares are the consequence of increased rent seeking (i.e., the top 1% increase their slice of the pie, while not increasing the size of the pie overall), while others have opined that the returns to innovation and highly skilled labor have increased for poorly understood reasons.  Many such as Joseph Stiglitz and Michael Roberts in an earlier post have pointed out that Hawaii’s economy has many uncompetitive industries such as electricity production and shipping.  However, this would suggest higher top income shares than elsewhere, not lower.  Another possibility could be that the industrial composition of Hawaii excludes much of finance, technology and biotech.  This suggests that a relative lack of highly specialized professionals in Hawaii may be responsible.

--Jonathan Page and Timothy Halliday

*Both the IRS and HIDOT report the number of returns by adjusted gross income (AGI) group (e.g., less than $1,000; $5,000 to $10,000; etc.). The total AGI reported for each group of returns can then be used to construct the proportion of total income a given group of families can claim (we use the total AGI reported instead of the income component from the national accounts). Income brackets reported by the tax agencies do not perfectly align with the percentage groups used by Piketty. For example, in 2005 those returns with AGI below $50,000 comprised 72.85% of all returns in Hawaii. To overcome this issue, we follow the procedure in Piketty and Saez (2003) to estimate incomes at the top of the income distribution. 

**HIDOT data covers 1958-2005, IRS data are used for 2006-2011.


Autor, D. H.; Dorn, D. & Hanson, G. H. 2013. "The China Syndrome: Local Labor Market Effects of Import Competition in the United States," American Economic Review, 103(6), 2121-68.

Corak, M. 2013. “Income Inequality, Equality of Opportunity, and Intergenerational Mobility,” Journal of Economic Perspectives 27, 79-102.

Piketty, T. & Saez, E. 2003. “Income Inequality in the United States, 1913-1998.” Quarterly Journal of Economics 118 (February), 1 – 39.

Page: 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12