There was an error in this gadget

Thursday, September 15, 2011

Unemployed Older Americans: An Annotated Bibliography (Draft from 2010)

Did this last year as a class assignment. Been meaning to post it. Not sure what the highlighted entries were about.

One of the findings of my lit review was that if you're between the ages 45 and 60 and loose your high-paying job, you likely face having to let go of your career and start a new one. Chances are very high that you will not get back into that previous career nor make the kind of money you had made.


Maestas, Nicole and Julie Zissimopoulos (2010)
            How Longer Work Lives Ease the Crunch of Population Aging.
            Journal of Economic Perspectives. 2010. Winter, 139-160.

Keywords: Labor Force Participation Rates, Economic Dependency Ratio, HRS.

Looks at future population distribution in terms of labor force participation (LFP) and theorizes that if LFP remains steady from 2010 to 2030, then economic dependency ratios (EDR) should also, logically, remain (relatively) stable (basically, if all barns are red, then all red barns are barns). Authors frame the dependency ratio as a measure of consumption needs relative to an economy’s productive capacity.  Using both HRS, CPS, and BLS data, authors challenge BLS projections that EDR will jump as boomers transition to retirement because this generation will largely prefer to remain working during typical retirement years. Discussion of Supply-side and demand-side factors in LFP for older Americans. Validates the presumption that many retirement-able Americans want to continue working but cannot find work.  Also contains comparative graphs of U.S. Population by Age and Labor Force Status in 1990, 2010, and 2030. showing the population momentum effect; Labor Force Participation Rates by Sex and Age, 1950–2000, and Projected 2010–2030;  and LFP comparisons of US with other OECD Countries (suggesting the US situation is at best moderate and not a looming disaster as is facing some EU countries, Japan and Korea).


Kim, Jeungkun.
Early Retirement in the Three Types of Welfare States. Research on Aging. V. 31.
No. 5. September 2009 520-548.

Keywords: Early retirement, Policy, OECD, TSCS[1].

In EU and other OECD countries, early retirement exacerbates fiscal debt problems (retirement ages in France, for example, are 60) and will increase labor shortages in some sectors. Because of this condition, EU researchers are looking for ways to encourage older workers to work longer whenever possible. The author builds on Esping-Andersen’s (1990) typology of welfare states by using (were social-democratic welfare states, conservative states, and liberal states) LFPR of 55-64 aged OECD men and women to compare state-level differences. Uses only aggregate data at country levels, pooled (time-series) and cross-sectional data from 1984-2001 of OEDCD members. Some findings were social-democratic welfare states with highest rates of absenteeism. Comparatively large social service employment seems to be prevalent in the social welfare states by maximizing care for the aged population. Differences in LFP rates differ also between men and women. Author suggests that policies to change early retirement patterns have different outcomes based on the welfare-market mix as well as gender.


Benítez-Silva, Hugo and Frank Heiland
Early claiming of social security benefits and labour supply behaviour
of older Americans. Applied Economics, 2008, 40, 2969–2985

Investigates the relationship between Social Security claiming behavior, monthly benefit claim reduction factors, and the decision to remain or reenter the labor force.  Taking the first five waves of HRS data, authors apply a simultaneous equation model to predict the exit hazard for individuals of early retirement benefit claiming (modeling the simultaneity of choosing to receive benefits and continue working.  Continuing with assumptions from earlier work, authors assume that subjects do not fully appreciate the consequences and the nuances of claiming early benefits on the Actuarial Reduction Factor applied to their long-term benefits. The data set comprises HRS subjects below the first month beyond turning 65 and left-censored cases in their data set. Most interesting finding of all is that individuals may be deciding to continue working and claiming early benefits in order to contribute more to their Social Security or(it follows, although not from the authors’)  other retirement instruments.

Note: Several earlier and longer versions exist freely accessible on the web. Right censored data refers to an event of interest occurred to the right (later) of some cut-off time. In this study such cases would have been excluded. The event in this study was the early claiming of benefits and remaining in the labor force.

Kemmerling,Achim and Miriam Hartlapp.
When a solution becomes the problem: the causes of policy reversal on early exit
from the labour force. Journal of European Social Policy. Vol 18(4): 381–394.

Keywords: EU, retirement policy, unit costs of labor.

Authors compare how two EU countries, Netherlands and Belgium, reversed the trend of about three decades of Early Exit Policies (EEPs) for older workers. Using the actor-agency framework, it was found that ruling party ideology (most often politically motivated) and the nature of trade union representation determined the speed nature of the revision of the mix of policies making up EEPs. The term ‘Active ageing’ policies are now seen modifying EEPs, as evidenced in the Lisbon Strategy. [2]. EEPs have come in a variety of schemes and devices and cannot be categorized under a single concept as lowered regular retirement ages as disability payments and long-term unemployment benefits for older workers are also examples of an (institutional) EEP. Authors make the astute observation that in such a case a policy reversal can only be framed as a policy outcome. To test their hypothesis, authors used simple correlation analysis. Authors showed how EEPs were politically popular in the 1970’s and seemed to alleviate labor supply pressures. By 1983, however, the correlation of unemployment and employment rates for 55-64 aged men trended slightly negative; and, by 2003 the correlation was more than twice that. They claimed that EEPs are now more of a cause of higher unemployment than providing relief on higher labor market supply  OECD countries with high unemployment were also found to have high EEP rates (eg., France). A corresponding finding was that of a positive relationship between low employment rates and EEPs for 55-64 year olds and a generally higher unit cost of labor.

Lippmannn, S.
Rethinking risk in the new economy: Age and cohort effects on unemployment
and re-employment. Human Relations. Volume 61(9): 1259–1292.

Keywords: demographic models
Study from an institutional perspective. Assumption that structural changes (due to the new economy) appear to make all workers liable for job loss, especially older workers (as they are conventionally thought to be the most vulnerable). Author uses cross sectional, pooled data [3] (20+ years worth) to investigate how older workers recover from displacement and reenter new jobs. Statistical modeling showed that showed birth cohort was a reliable predictor for post-displacement employment outcomes more so than specific skills or knowledge. The principle finding was that older workers who were more functionally adjusted to, or suited for, a labor market in which the worker-employer social contract is weak are more adept at reentry post job loss.

Note:  technical and econometric modeling, tables.


Benítez-Silva, Hugo and Frank Heiland
            The Social Security Earnings Test and Work Incentives
Journal of Policy Analysis and Management, Vol. 26, No. 3, 527–555.

Keywords:  HRS, Social Security, Earnings Test, Actuarial Adjustment Factor, Policy
Policy angle advocating caution with regard to changes to Social Security. Examines the benefit incentives of Social Security for early to normal retirement aged people. Shows how the LFPR among older Americans is increasing along with substantial rate of early retirement claims (70.6 % for men, 75.5 % for women in 2006). Authors claim that policy makers and researchers are not adequately addressing the implications of working while receiving Social Security income and with HRS data analysis, show that workers who receive early benefits and continue working are not sufficiently aware of the long term consequences to their retirement incomes. Additionally, authors claim that the Earnings Test is distortionary and costly to administer. Convincingly argued that the length of time since claiming early benefits positively affects the decision (or preference) to stay in the labor force. Introduce a life-cycle, structural model (assuming an open system with subsystems as the source for dimensions). They identified a basic eGovernment problem at the time of writing in that the Social Security website does not provide any calculation of the actuarial reduction of lifetime benefits due to a person’s early retirement. Also found that in the first five waves of HRS data for ages 62 and over, less than 2 percent were not working to do unemployment.

Note: Moderately complex statistical analysis with predictive modeling, tables.

Eibner, Christine (Rand), Alice M Zawacki  (U.S. Census Bureau), Elaine M. Zimmerman (U.S. Department of Labor).
HEALTH INSURANCE, 2000-2004. Center for Economic Studies (CES)
Research paper.

Keywords; retiree health insurance (RHI), Medicare Modernization Act (MMA)

Authors attempted identifying and separating the factors influencing employer provisioning of Retiree Health Insurance (RHI). Assuming that 21% of the labor force will be 55 and over by 2014, analysis of retirement and health insurance factors is critical to modeling future labor trends. Using multiple data sources and a multivariate regression approach, authors looked at the character and degree of influence of labor markets, unemployment rates, union presence, employment sector, firm size, economic sector, and the Medicare Modernization Act (MMA --the a prescription drug benefit to Medicare Part D, January of 2006). Findings mostly confirmed the work of other researchers. Fewer firms are offering RHI (a trend beginning around 1990) and those which are require higher financial contributions from retirees.

Note: 37 pages, several tables, complex econometric models.

Maestas, Nicole.
Cohort Differences in Retirement Expectations and Realizations. From Chapter 2
of Redefining Retirement: How Will Boomers Fare? Brigitte Madrian, Olivia S.
Mitchell, and Beth J. Soldo, editors. Oxford University Press. Oxford. 2007.

Keywords:  HRS, Dynamic Retirement, cross-cohort analysis.

Author uses HRS data for cross-cohort analysis of retirement, labor force attachment, and post retirement employment. ANOVA and multivariate regression models show that all cohorts, especially boomers, have greater labor force attachment. However, this cannot be explained solely due to health (ability to work) or socio-economic reasons. Findings suggest that intangible factors such as self-worth, work ethic, or other yet identified reasons help explain it. HRS Cohorts were categorized as the Children of the Depression Age (1926–30), the original HRS (born 1931–41), War Babies (born 1942–47), and Early Boomers (born 1948–53). Findings of other researchers (perhaps using the HRS, as well) were validated. Labor force participation trends upward among latter cohorts. More than one single factor contributes to this as typical explanations usually consist of the relaxation of mandatory retirement in 1986 and less attractive financial incentives. Retirement, per se, is a multi-faceted somewhat dynamic condition as older workers do not completely withdraw from labor force following full-time employment or careers. For 50 to 75% of HRS cases, bridge employment (part-time or different work from that of a career) seems to be what retirement means. Author finds that Early Boomers generally posses higher education, greater ethnical diversity, have higher mean incomes and higher total net wealth. Notably, the Early Boomers have lower rates of defined benefit (DB) pension plans. This cohort anticipates retiring at age 64 on average and have higher probabilities of working full-time up to age 65. They also underestimate their chances of reaching age 75. 

Note: Moderately complex statistical analysis with predictive modeling, tables.

Coile, Courtney C. ,  Phillip B. Levine
Labor Market Shocks and Retirement: Do Government Programs Matter?
NBER Working Paper No. 12559. October 2006

Keywords:  Unemployment Insurance, Retirement decision.
Quantitative modeling investigating how Unemployment Insurance and Social Security (SS) affect older workers response to negative labor market shocks (increased unemployment rates). Using a cross-sectional data set drawn from the CPS and the HRS,  the authors concluded that higher unemployment rates will lead to increased retirement decisions for those workers eligible for Social Security. Authors conjecture that SS becomes a more attractive form of UI for older workers. This is because of the fact that UI’s purpose was meant for consumption-smoothing to balance income loss due to unemployment and to help an unemployed worker find a new job. Older workers, once unemployed reason (rationally) that they have less to benefit from finding a new job (due to fewer working years left). UI benefits also have limited duration while SS benefits can be received over several years. This makes SS preferred over UI.

Maestas and Xiaoyan Li.
Discouraged Workers? Job Search Outcomes of Older Workers. Working Paper (WP 2006-133). Michigan Retirement Research Center, University of Michigan.
October 2006.

Keywords:  HRS, Social Security, Earnings Test, Actuarial Adjustment Factor, Policy

With HRS data, authors studied employment transition rates for 1992, 1998 and 2004 among age bands of 51-56, 57-61, 62-67, and 68-72. Largely focused on econometric modeling of a reservation wage explanation for unemployment among older workers. Study verifies the growth of retirement aged workers seeking reentry to the work force and that their success (or failure) of attaining employment is not well documented. Job attainment and reentry transition rates for older job seekers appear to be much lower than for younger cohorts in similar unemployment conditions. Older job seekers appeared to have a lower wage demands yet still experience very low attainment rates. Authors speculate that this rate might confirm statistical evidence of age discrimination. Older job seekers tend more often to become discouraged workers (by strict definition). Boomers in this cut of the HRS appear to have a stronger attachment to the labor force than previous cohorts. Therefore, it should be expected that unemployment will be a far more prevalent problem for them than as in earlier cohorts.



Chan,Sewin and Ann H. Stevens.
How Does Job Loss Affect the Timing of Retirement?. Contributions to
Economic Analysis & Policy. Volume 3, Issue 1 2004 Article 5.

Keywords:  HRS, Retirement decisions

Models the relationship between job loss and retirement decisions to explain in part unemployment among older workers. The assumption was that retirement could incentivize leaving the workforce early, or is chosen in lieu of being able to find a replacement job. It was found that the effect of job loss was dramatically negative on wages and pension wealth. Deep initial wage rate losses, averaging a whopping 51% and remain at 23% below expected wages six years thereafter (the cohort data was six years in total). Authors struggle somewhat to define retirement (full, partial, is it one event or are there such things as retirements in which a person drops out and back into the labor force, repeatedly changing their self-reported status). Of further note, the HRS data showed dramatic deviations in Average Annual Earnings, Average Total Assets, and Average Pension Gain in the data.

Holtz-Eakin, Douglas, Mary E. Lovely and Mehmet S. Tosun.
Generational conflict, fiscal policy, and economic growth. Journal of Macroeconomics. 26. 2004.

Keywords: dependency ratios, growth, econometric models, fiscal policy.

Increased dependency ratios over the next 50 years will put pressure on fiscal policies concerning education. That is, the more elderly we have in our population, the less willing they will be supporting taxes for education. Authors hypothesize that this demographic will have a preference for less spending, which will indirectly reduce capital investment of education. This demographic transition will lead to reduced output along side larger capital stock. The impact of the aging population and lopsided dependency ratios impacts long term growth but present fiscal policies are not addressing this transition. The author’s models assume a closed economy (authors admit that in reality capital flows internationally to the highest return). Authors are influenced endogenous growth theory (See Paul Romer’s work). Endogenous growth theory basically explains economic growth rates are based on how fiscal policy affects R & D (patents the monopolies resulting from them) and education (long term intellectual/knowledge capital investment). This is in contrast to exogenous growth (See Solow’s 1987 nobel prize work) --aka the neo-classical growth model. Years covered?
Note: Highly technical econometric modeling, graphics and tables.

TRANSITION. Ageing International, Summer 2004, Vol. 29, No. 3, pp. 281-308.

Keywords: Microeconomic forces, employer preference, policy.

Focus of research on Australian workers. Asserts that OECD country policies meant to increase older workers are not working as intended. Author uses previous survey data (1999) from Australian human resource managers. Findings (descriptive) were indicative of ambiguity of older workers among respondents. That is, the potential value to a firm of training, retaining or rehiring older workers is not realized. Cost of employment of a younger worker, and the conventionally assumed better long-term investment, was a major factor. Additionally, older workers themselves often consider themselves are less employable. Author suggests public policies are too simplistic and do not positively influence change among government, employers or the older workers themselves.

Note: Descriptive statistics, graphics and tables.


Borsch-Supan, Axel
Labor Market Effects of Population Aging. LABOUR 17 (Special Issue).  5-44

Keywords: Predictive modeling, structural effects.

Analysis of the effects of population aging on the labor market and determines their broad implications for public policy. It takes Germany as an example but findings should apply to the other large economies like the US. Findings suggest that higher capital intensity and efficiency will not offset the entitlement burden of the large retirement age cohorts in the near future. Additionally, demand and supply structures shall change in accordance with the coming demographic shift. Authors predict that employment patterns will change also and generally require an increase in labor mobility to accommodate such structural shocks.

Morris, David, Tony Mallier
Employment of Older People in the European Union. LABOUR 17 (4) 623–648.

Keywords: European Union, Bridge employment, Self employment

Discussion of what the authors term the alternative employment modes (being, mostly, part-time and self-employment) of older workers in the EU. These alternative modes are examples of bridge employment, a transitional state into full retirement. These changes became prevalent in the EU due to the climbing rates of demographic ageing (similar to that of the US). Models used Eurostat Labour Force Survey data. There are (at the time the paper was written) 15 separate labor markets in the EU. Three of those appear to show that such alternative employment modes are due to fewer full-time jobs.  Yet, evidence exists showing that older workers prefer the flexibility of such alternative employment modes, with self-employment being very popular among older men. The demand in EU labor markets appear to favor such alternative modes as well.

Knut Reed — Fredrik Haugen
Early Retirement and Economic Incentives: Evidence from a
Quasi-natural Experiment. LABOUR 17 (2) 203-228.



Benitez-Silva, Hugo.
Retirement Research at Boston College. December 2002. 38 pages.

Keywords: Job Search Behavior, HRS, Predictive modeling

            Claims to be the first dynamic model of job search behavior of older Americans. Author presented a model of job search using all available waves of HRS data. Findings showed that older Americans were actively engaged in the labor market when both employed or not. Job search behavior tended to be dependent on gender and previous work and health status. Basically a utility maximizing model. Some findings were that with longer life expectancies, better health and new opportunities created by info tech many currently employed were searching for bridge jobs (often part-time) for transitioning from work to full time retirement. Most cases in the HRS sample if in their fifties were found to be actively searching. Certain other expected factors such as health insurance or pension status influenced the choice to search for work between HRS interview periods.

Note: probit econometric modeling

Diamond, Peter A. and Peter R. Orszag

An Assessment of the Proposals of the President's Commission to Strengthen Social Security. NBER Working Paper No. 9097. August 2002.

Keywords: Social Security, Retirement decisions,

Foundational report for policy analysts on contemporary Social Security reforms, especially regarding privatization. Critically and fairly assesses the 2002 Bush Commission Social Security Proposals. Finds that of the three models proposed by the commission, models 2 and 3 hold the greatest merit for reaching the goals of sustainable, long-term actuarial balance of the program. Authors show how a scaled-back social security system with benefit reductions would look like given contemporary assumptions of demographics, life expectancies, modest economic growth, among other things. Diversion of earmarked social security revenues into individual retirement accounts exacerbate the program’s legislative mission. Greatest value of the report was the introduction of a new baseline with which to evaluate Social Security reform plans.

Note: About half of the paper’s 62 pages are made up of tables. Lay-readable critique of the 2002 Bush Commission Social Security Proposals.

Spiezia, V. The Greying population: A wasted human capital or just a
social liability? International Labor Review, 141, 2002. 20-44.       

Keywords: policy recommendations, OECD, dependency ratios, capabilities of older workers.

            Though pertaining primarily to non-US, the author states several generalizable concerns about older workers in developed economies. Primary focus with dependency ratios of younger (replacement workers) to older workers. Identifies pressure on the younger labor force with an analysis of labor market effects of aging and provides policy recommendations to the OECD (mostly relevant to EU countries). Several tables concerning participation rates from 1970 to 2000 compared across countries including the US.  [4] At the time of the study, displacement rates for older workers in OECD countries were rising (author terms this as the “hiring intensity” with a 1:7 ratio between the population of 46-54 to 15-24 year olds). Author sees this as largely due to age discrimination as convention has it that older workers are less productive. However, his findings suggest a rubric tailoring certain age-appropriate categories of work and presents a policy framework to so accommodate.

Note: Sandell Grant Program paper for Junior Scholars in Retirement Research.
Simple descriptive statistics, some econometric modeling, tables.

Chan, S. and Ann Huff Stevens.
Job Loss and Employment Patterns of Older Workers. 2001.
Journal of Labor Economics, 19:2, pp. 484-521.

Keywords: HRS, postdisplacement employment models.

Authors used HRS data for re-employment trends of workers 50 and above who were involuntarily unemployed. Simultaneous equation hazard models were used for estimating subsequent re-employment at 10 years after a job loss. Authors show that involuntary unemployment of workers 50 and above results in longer-term unemployment and lower probabilities of re-employment in comparison to younger cohorts. Model estimations of re-employment rates were about 70%-75% within 2 years following an initial job loss along with reduced earnings (citing Jacobson et al [1993] that earnings are reduced up to 25% and up to six years in postdiscplacement).  Such re-entry to the labor force employment is termed postdisplacement jobs. Such re-entry jobs are characteristically brief with increasing probabilities of exiting the labor force permanently. Thus, authors demonstrate that for older workers standard econometric job search and reemployment models apply but retirement models must also be included in analysis. Rates for displaced workers in their 60s were even lower. Lit review documents that from 1981 to 1993 job loss rates for 55 were on a riding trend.

Steuerle, Eugene and Adam Carasso.
A Prediction: Older Individuals Will Work More in the Future. Straight Talk on
 Social Security and Retirement Policy. No. 32. March 30, 2001. Urban Institute.

Keywords: Social Security, labor force attachment,

Short policy memo (about two pages) illustrative of the policy stream debate that Social Security solvency may necessitate greater, more sustainable rates of labor force participation among older workers. Written in 2001, authors contend that increased rates of the retirement aged workforce due to demand pressure. Authors predicted this demand to remain robust in the near future.

Williamson, John B.  and Tay K. McNamara.
Why Some Workers Remain in the Labor Force Beyond the Typical Age of
Retirement.  November 2001.Center for Retirement Research at Boston College

Keywords: HRS data, predictive models, retirement decisions

Models the effects of age, gender, race, and non-work income on retirement decisions in 1998 HRS data. Authors reported their study filled a gap in the contemporary literature due to the fact that it models (using a binomial log form) remaining in the labor force during conventional retirement ages (60 to 80 years of age). Literature review shows the trend of post-retirement labor participation to be a recent one, beginning in the mid 1980’s to 1990’s. Findings show that the decision to retire to be very complex, dependent on multiple factors (education, race, gender, non-work income, age, functional capabilities, health, among others). In part, authors address their findings to policy makers who want to keep older workers in the labor force to include sub-groupings of worker classes in policy proposals.


Benıtez-Silva, Hugo. Micro Determinants of Labor Force Status Among
Older Americans. Draft version. SUNY at Stony Brook.

Keywords: Reverse transition, self-employment, HRS.

Author takes over 20k cases from three waves of HRS data and constructs a probabilistic regression model (multi-nominal logit model/Maximum Likelihood). Terms the action of continuing to work, or going back to work, after retirement as “reverse transition”. Looks for microeconomic variables influencing the decision to engage in reverse transition. Declared that older Americans were able to find work easily (initial draft written in 1998 and updated by 2000, during a economic bubble period).
Finds strong evidence showing that at 62 to 65 the status of health insurance lowers the probability of reverse transition, especially for Medicare recipients. Models the likelihood of reverse transition between men and women. Several interesting findings among which was the tendency of moving to self-employment for whites of higher personal income and education; while total net wealth (personal income being inclusive) negatively related to re-employment. Suggests retirement age personal/employment income as a reasonable proxy for job skills/employability.

Note:  Very technical econometrics, tables and graphs.

Burtless, Gary and Joseph F. Quinn
Retirement Trends and Policies to Encourage Work Among Older Americans. Brookings Institution, Boston College. Economics Department Working Papers in Economics. Boston College Year 2000. the Boston College Center for Retirement Research.

Keywords: Policy analysis, Social Security, post-retirement age employment incentives.

Authors present a critical and thorough policy analysis on whether to and how to encourage retirement past the NRA. Considers whether such policy changes are needed since socioeconomic factors (general increase in economic wealth and worker productivity), life expectancy and improved health are shaping post-retirement age work force trends well ahead of policy designs. Also explains how Social Security policy changes contributed to the trend of early retirement. Shows how this was historically unusual and limited to only a short period (early 1960’s to mid 80’s). Authors make interesting assumption that job creation in the American economy is strong and sustainable (perhaps due to the time they wrote their analysis), with the corollary that anyone who wants a job should be able to find one (irrespective of age).

Costa, Dora.
Has The Trend Toward Early Retirement Reversed? Prepared for presentation at
The First Annual Joint Conference for the Retirement Research Consortium
“New Developments in Retirement Research”. May 20-21, 1999

Keywords: retirement history,

An economic historian claims that higher retirement trends for men an historic trend and unlikely to reverse in the near future (if at all). His thesis is based on consistent declining labor force participation during the periods of 1948 to 1985 followed by only a modest rise since 1985 (to 1999 when this material was presented). The author lists the seven most frequent explanations of why retirement trends were supposedly reversing, upping the labor force participation of older men. Among these were the change in defined benefit plans, fewer private pensions, changes in the earning test, l better health and greater longevity, and simple life-style choice including the popularity of  bridge-jobs. But men have chosen retirement mostly because of higher incomes and wealth and the presence of, what should be called, the retirement industry, which makes retirement easier to arrange.

Note: Charts, some tables.


Blau, David M.
Labor Force Dynamics of Older Married Couples. Journal of Labor Economics, 1998, 16, 3, 595-629. University of Chicago.

Keywords: Married couple, Retirement History Survey[5], Dual entitlement rule

The labor force status and labor force attachment of one partner on the other in a married couple influenced the retirement decisions of both. The differences appeared related not to financial factors as much as preferences for leisure. Also found what that the dual entitlement provision of Social Security[6]  negatively influenced labor participation by wives and had a small positive effect on husbands’ choice for retirement. Most husbands appeared to leave the labor force prior to their spouse while retirement order appeared very dependent on cohort. Savings behavior appeared not to be modeled and was exogenous although private pension income was not modeled also a dummy was included. Most modeled variables lagged from one time-state to the other. The existence of pension coverage appeared to increase labor force exit rates. The effect on the wife’s entry rate from (or attachment to) the labor force was negligible whereas the husband’s exit rate was 2x greater if the spouse was not employed.

Note1: need to go back to this one   .

Note2: Moderately technical econometric modeling, tables and graphs.

Blau, David M.
Social security and the labor supply of older married couples.
Labour Economics 4 (1997) 373-418.


[1] Time Series Cross Sectional.
[2] Lisbon Strategy was established in March 2000 by the European Council in Lisbon was a set of goals to make the EU the world’s leading economy. 
[3] “A cross-sectional data set consists of a sample of individuals, households, firms, cities, states, countries, or any other micro- or macroeconomic unit taken at a given point in time. Sometimes the data on all units do not correspond to precisely the sametime period.”  INTERNATIONAL ENCYCLOPEDIA OF THE SOCIAL SCIENCES, 2ND EDITION, p  367-69.
[4] NB: Studies involving non-US data, if relevant, will be included. (note to self: page 19 notes a distinction s/b made between factors influencing labor force participation and the demand of labor. Page 16 aging seems to stimulate innovation rather than reduce it Cutler et al 1990).
[5] Retirement History Longitudinal Survey Series. Investigated by the SSA. Contains Six biennial survey samples by the US Census Bureau from 1969 to 1979.
[6] Dual income couples are treated as a single beneficiary. The Social Security retirement earned as a worker is subtracted from spousal benefit SS and only the difference is given as a spousal benefit. Haltzel, L. Social Security: The Government Pension Offset (GPO). CRS Report RL32453. January 11, 2008.

Wednesday, August 3, 2011

Foreign Aid: Is it effective?

It is difficult to judge whether foreign aid has been an effective strategy for encouraging “development” as there are somewhat hazardous definitions and assumptions floating around both concepts (aid and development). To determine whether the effects of foreign, or even domestic, aid leads to desirable outcomes depends on the circumstances to which that aid was applied and how the outcomes were defined and were to be measured.

The World Bank takes a round-a-bout way of defining “development” in terms of economic growth. Economic growth itself is related to a general increase in a society’s wealth. Wealth, in this sense, would imply a net decrease in poverty in an economy to its capacity to provide goods and services that directly contribute to, more-or-less, westernized ideals of a quality life (access to effective health care, sanitation, potable water, safety, education and economic opportunities, democracy and so on). [i] Thus, development is often associated directly to reductions in poverty and its associated chronic problems (health, sanitation, even an enabler of democracy).

But economic growth also has less desirable consequences. Growth can amplify class and economic disparities, lead to unmanageable urbanization areas, slums, environmental degradation, have no clear correlation to democratic forms of government (China is a stellar example) and, ironically, economic instability (take for example Greece, Spain, Portugal at the present time).

In general, thanks to economists as Banerjee and Duflo, the domain of development economics is finally trying to address its failure in alleviating the real problems human beings face. This is largely due to its insistence upon its own theoretical purity. Too often, economists are driven by ideology and resist challenging their honored assumptions (embodied in their models) by the insights from other sciences (psychology, anthropology, for instance). The work of Banerjee and Duflo evidences an effective challenge to ideologically driven economic assumptions by, in some ways, relating to us lessons that I recall were taught by the work of Marvin Harris several years ago. In his approachable, simple book, Cows, Pigs, Wars, and Witches, written in 1975, he showed to the lay public that what we assume to be irrational or bizarre behavior in an alien culture (and, most certainly, the poverty Banerjee and Duflo work with is indeed an alien world to most of us in America) has rational reasons to exist to the people in that cultural milieu.

For instance, Jeffrey Sachs believes foreign aid, if applied rationally and in large scale, can be effective in alleviating poverty. He presents a list of characteristics that aid programs should follow that are very optimistic and likely too unrealistic and impractical in today’s environment. Notably, Sachs believes aid should be centered on measurable outputs and less upon abstract ideals as democracy and economic growth.  There is value in Sachs’ arguments however in that he emphasizes understandable, readily available technologically adept inputs, achievable targeted outputs, adequate financing from coalitions of funders. We should take Sachs’ argument with moderate seriousness because his approach seems based upon the flawed assumption of economic rationality (a least from the point of the intervening agency; the happy-path assumption that if we apply simply technology smartly, we get smart results, is deceptive).

Representative of more-or-less Libertarian inspired economic theories, espousing that aid causes more harm than good, would be the arguments of George Ayittey. He rightfully criticizes the premise of foreign aid by pointing to disappointing, conflicting results in Africa. He claims aid is motivated less by practical, realistic planning but rather from emotionally (meaning, irrational) biased motivations on the part of the developed countries. Aid invites corruption, waste,  creates dependency and discourages innovation and empowerment for those to whom the aid is directed. But this kind of thinking is even more flawed than that of Sachs as it supposes the poor in foreign cultures are motivated by the same economic desires and goals as Ayittey himself.

Unfortunately, none of these authors define complete, realistic models that seem capable of providing a roadmap to a future of significantly reduced human suffering (which is the whole point of intervening with aid, or at least it should be). Banerjee and Duflo however have presented findings are based on very creative, informative experiments and meaningful data (or so it appears). Additionally, their sensitivity and ability to describe and understand “irrational behavior” is a very important example of critical thinking necessary for addressing this problem.

Banerjee and Duflo argue convincingly that common evaluation models for the effectiveness of aid is inadequate, or biased (if it is evaluated at all) and must be improved. The value of their work (aside from its successes, however limited in scale they appear), is to show that the so-called irrationality and short-sightedness of the poor that frustrates aid programs, or makes them appear to be so ineffective as to be ludicrous, has more to do with the blinders of culturally-biased assumptions of the aid granting countries.[ii] Banerjee and Duflo demonstrate an effective approach to researching and gaining understanding of how to address specific, small-scale improvement opportunities. They are also collecting valuable data.

Ignoring the potential ethical questions raised by critics (pointed out by Easterly in his NYR article) who observe, perhaps rightly, that aid is motivated by paternalistic assumptions, or even some kind of left-over colonialist collective guilt, of the aid granting countries, a interesting thought experiment is to imagine what the implications would be of a large scale replication of Banerjee and Duflo’s five lessons to “development” (leading to a significant alleviation of poverty).

For this experiment, suppose there are 1.5 billion “poor” in the world. Imagine if 30% of the world’s poor were effectively lifted from poverty (by whatever definition you have) in one or two generations (20-40 years). Now the world has about 450 million healthy, educated, intelligent people. Assume they have benefited from western-style prenatal care and their educational level has a mean of primary and secondary education (maybe up to our notions of an 8th grade level).

Have our models of development outcomes aimed at alleviated poverty given any thought as to what happens next? What are those healthier, smarter people going to be doing with themselves? Will there be a social, economic, environmental infrastructure capable of carrying this new population? To assume that all of the younger, smarter, ambitious would aspire to stay in their rural or small city settings to be teachers and social workers is hardly realistic. Would they instead be compelled to migrate to urban areas to seek economic opportunity? Would those urban areas be ready for this new migration? The larger scale cities hardly seem able to do that now.

The point here is that models of development need to take a whole system-of-systems in approach in its design, its measurement definitions, and not stop their models at the point of desired outcomes. In the Evaluation Logic Model [iii]sense, Outcomes lead to new situations with new complications and potential problems to solve. There are unending chains of Situation-Target-Outcomes in development aid. Planning and execution has to be informed such that aid does not lead to unintended consequences. Therefore, it seems an approach like Banerjee and Duflo’s coupled with some kind of systems thinking methodology (approaches like Peter Checkland, Gregory Bateson) could be developed as our next step.

[i] World Bank. Beyond Economic Growth: Student Book.

[ii] Easterly, William. Measuring How and Why Aid Works--Or Doesn't. Wall Street Journal - Eastern Edition. 4/30/2011. Vol.257,Iss.100;p.C5.

[iii] Program Development and Evaluation, Univ. of Wisconsin, Retrieved 20110623.

Tuesday, August 2, 2011

The Death of Vishnu, by Manil Suri. W. W. Norton & Company. January 2001

This was Manil Suri’s first novel, written in 2001. He is an applied mathematics professor at the University of Maryland, Baltimore County. [1] He was the author of a few short stories prior to his first novel. When this book was published, the author had been away from India for about 20 years (he had in fact come to the US as a student at Carnige Mellon when he was 20). However, once it was published in India, some prominent critics there called it an “Indian novel”.  The novel takes place in contemporary Mumbai and is based upon some of the people the author knew of in his apartment building while growing up in Mumbai.

The stage of the story centers on the lives of people in an apartment building. The protagonist, Vishnu, inhabits a stair landing between the ground and first floors. He is very poor, somewhat of an alcoholic, middle aged, and lay dying on his landing. Vishnu gained this prized spot in the building (prized due to his standing or caste) by purchasing the right from the previous occupant. Additionally, in exchange for occupying on the landing as his residence in the building, he does odd jobs for some of the families (such as standing in line to get milk in the morning, washing dishes, and so on).

Life in the building goes on as he lay dying. There are two Hindu families living on the first floor, a Muslim family on the second, and a widower who rarely comes out of his apartment on the third floor. The two families on the first floor have an ongoing tension, rivalry, and bickering going on between the wives over several of things. For example, their shared kitchen (on the same floor) is a place of a sustained passive-aggressive arms race. Accusations fly from one woman against the other for assumedly taking more than her share of water ration from the cistern on the roof. This could bring on a reprisal in the guise of “borrowing” the other family’s ghee (clarified butter) or other, and so on.

As it slowly dawns on these women that Vishnu is ill (the novel begins with one of the women delivering Vishnu’s morning tea and finds him comatose and having soiled himself), they begin fighting over which family should take responsibility for doing something. This becomes an ever more volatile issue between them as the day goes on, both demanding of the other to pay for an ambulance to take him to the hospital. At one point, the two husbands, neither of whom have much control over their wives and generally keep out of the way of the conflicts and from each other, decide to arrange for an ambulance between each other without their wives’ knowledge or approval. Once the ambulance arrives, neither wife agrees to pay the ambulance attendant until after much arguing and struggle. Once the ambulance has been given their fee, he demands someone sign for the responsibility for Vishnu’s hospital bill. Neither family was prepared to go that far and the ambulance left, with Vishnu at that point either near death or having had actually died.

Much else of the narrative takes place in Vishnu’s mind as he fades in and out of consciousness and his memories of childhood, his youth, scenes of his lost love (who was a prostitute). Vishnu never married and had no immediate family in the sprawling city, his relatives living several hours away by train in the interior of India. Eventually, his soul begins to leave his body and he is able to hear the voices of the insects crawling around the stairway and within the walls, able to see the lives of the other residents go about their day, all the while continuing his fading from the present into the past. As he “climbs” the stairs, compelled by some urge to go to the roof of the building, be begins to believe himself to actually be the final incarnation of Vishnu, rather than being dead (which does not seem to occur to him at all). As he ascends to the top stairs, his visions become more and more laden with imagery of Hindu mythology until, at last, he enters the home of all the gods who are all waiting to greet him. His journey then rapidly takes him to the close of his story, into the company of Krisna who sits in a beautiful forest playing his characteristic flute. Vishnu, greatly puzzled at realizing he is not the incarnation of his namesake god, asks the boy ‘What now?’ to which is replied, ‘Why, you rest and then go back, of course’! Vishnu is then finally at peace after a long, very difficult life of poverty, struggle, and humiliating death (though appropriate for his station in life).

One of my favorite characters of the novel was Mr. Jalal. He was the husband of the Muslim family from the second floor. His story within the novel was significant as it became something of a flashpoint in one of the many subplots. Jalal was a skeptical, intellectual, and critical man, well read in western and eastern philosophies and religions. Mostly an agnostic (while other characters, including his wife, were obsessively superstitious and religious), he was at the time of the novel dealing with a spiritual crisis of his own. He was seeking for himself some kind of religious experience, or feeling, something that did not appeal to the intellect or the emotions but to the soul, and experience which he felt he needed but could not find. He believed the key to the religious experience was suffering and he tried several different methods to achieve the requisite amount. However, he was basically afraid of pain (‘why must pain be so painful’, he said to himself at one point) which further complicated his journey to arrive at a religious experience.

One of the acts of suffering Jalal attempted was to lie down with Vishnu on the landing, perhaps as an act of contrition to God (or gods) to be at a level of one so lowly as Vishnu, or perhaps to contract the disease that the residents assumed had had made Vishnu ill. Before doing this, however, the daughter of one of the first floor Hindu families and his son made their try at elopement. As the young woman went down stairs, she left an article of clothing on Vishnu as a keepsake.

Jalal had a fitful dream that he interpreted as an authentic religious vision of the Hindu god, Vishnu. He was awakened in the early morning of the next day by the other first floor family (rivals to the one whose daughter had run away with Jalal’s son). (Bear in mind Vishnu was likely very much dead at this point but the families were not all that interested it seems to verify this fact). Once awoken, he excitedly recounted his vision of Vishnu to them. He had, unfortunately, merely recounted a scene from the Bhagavad Gita that he was at the time not aware he had read years ago and had forgotten. But he was sure he had achieved his goal of a genuine religious experience and that he was now a prophet, of all things, Vishnu. It was his calling, he believed, to bring harmony to the two great religions of India, Islam and Hinduism.

When the vendors and others (all Hindus) who did not live in the building learned that the Hindu family’s daughter’s shawl had been left near Jalal, who was clearly crazy or drunk, and that his son was also missing, rumors quickly spread that the girl was a victim of foul play. A small mob formed, marched to the second floor, intending to extract justice. They beat Mrs. Jalal senseless and caused Mr. Jalal to fall from a balcony as he tried to escape.

What this novel portrays of a modern Indian city is quite exciting. All of the Hindu married couples had been arranged marriages. Each of these marriages started off on somewhat awkwardly but each had, it seemed, grew to be in love with each other. The daughter who attempted an elopement with Jalal’s son, was in fact having such a marriage arranged prior to the Jalal’s demise. It was interesting to see how she abandoned the younger Jalal later on the day of the elopement and after they left the city. Once she realized that they were to start a modest, humble life in a smaller city, she thought better of the arranged marriage: it was more attractive to her to be worshiped by the homely engineer her parents had arranged a marriage to. I had assumed that such marriages were mostly a thing of the past, not something in a modern, high-tech India (even pre-high-tech India of the 70’s and 80’s) would be common.

Next, the castes, or stations of life, the vendors and Vishnu had in comparison to the higher life of the building’s apartment residents was arresting: I had no familiarity of these separations of people other than a vague understanding of such a cultural condition. Further surprising was how the vendors who sold their wares to all the families with no apparent prejudice or ill will to the Jalal’s, and even in some cases had served that family for many years, would all succumb to being swept up in a religion-inspired violent act against them. 

There are also frequent scenes of how valuable space is to the lives of the characters. A stair case landing makes a legitimate home for someone. Some people of lower castes as Vishnu can take up residence on some of the lower stairs. The stair case itself serves as a metaphor for both spiritual journeys (for Vishnu certainly and then for Jalal, too, to a certain extent) and for caste, class, and rank (the higher floors were occupied by higher income families). Cramped, shortages of space also figures prominently to the two Hindu families whose rivalries and resentments emerge most often in the shared kitchen.

In one way, the novel could be a kind of social allegory for India. Vishnu embodies the interplay of earthly powerlessness but with a certain kind of spiritual power (through his progression from living to death). We see a vibrant, intense Indian city that is a crowded, noisy place. Your place in society, whether Hindi or not, makes up more of you than it would be made of in America. Tradition, religion, and superstitions (even spells, witchcraft and so on) are very important to many people (as it is here) that further defines the self in India for an Indian. This is a valuable, fascinating book and an impressive first novel.

[1] Manil Suri, Biography.  Retrieved 20110701.