Albert A. Okunade

Technology as a 'major driver' of health care costs: a cointegration analysis of the Newhouse conjecture

Per capita real income on the demand-side and technological change, proxied by total R&D and health R&D spending, on the supply-side are hypothesized as major drivers of per capita real health care expenditure in the US during the 1960-1997 period. The findings are robust to a battery of unit root and cointegration tests. They support the Newhouse [Journal of Economic Perspectives 6 (1992) 3] conjecture that technological change is a major escalator of health care expenditure and confirm a significant and stable long-run relationship among per capita real health care expenditure, per capita real income and broad-based R&D expenditures. Policy implications are noted.

DETERMINANTS OF CHARITABLE GIVING OF BUSINESS SCHOOL ALUMNI

This study investigated the propensity of Business School alumni to donate cash to the alma mater. A logit regression model was fitted to the survey response data from the 1955/56–1990/91 alumni of a large, Carnegie-classified, Doctoral University I public higher educational institution. The marginal probability of giving was found to be significantly related to factors such as time since graduation, major area of degree, willingness to recommend the university to others, household attributes and family ties to the alma mater, the number of other voluntary donors known, and the availability of matching gift accounts where alumni are employed. The logit estimates can be used to predict the likelihood of giving and for selecting fund-raising prospects more efficiently. Potentially rich future research directions are explored.

Unit root and cointegration tests: timeseries versus panel estimates for international health expenditure models

Classical regression estimates of the determinants of the OECD health expenditures are useful for policy formulation and evaluation. However, if the underlying timeseries data are not collectively stationary in levels, the estimated parameters are faulty and can misguide health policy. Until very recently, the crucial stationarity tests were ignored in a large number of studies on international comparisons. Stationarity (ADF, Phillips-Perron, IPS heterogeneous panel) and cointegration (Engle-Granger bivariate, Johansens multivariate) tests are conducted here using 1960–1997 health expenditures data (1998 CD ROM) of 19 OECD countries. It is found that extending the time series data length affects the order of integration and number of cointegrating vectors. However, it is arguable whether the order of integration decreases or increases as more observations are added for testing. The failure of the Johansen and Engle-Granger cointegration tests for most of the OECD countries cautions policy makers against reliance on earlier research findings that were based on unstable relationships among variables in the regression models. (This is not the case for the UK, Greece and Ireland; policy implications have been derived for the UK.) Consequently, data calibrated in growth rates may be more appropriate for investigating the long run relationships collectively in a panel of OECD health expenditure model specifications.

Determinants of Health Expenditure Growth of the OECD Countries: Jackknife Resampling Plan Estimates

Due to the lack of internal consistency across unit root and cointegration test methods for short time-series data, past research findings conflict on whether the OECD health expenditure data are stationary. Stationarity reasonably guarantees that the estimated OLS relationship is nonspurious. This paper departs from past investigations that applied asymptotic statistical tests of unit root to insufficient time-series lengths. Instead, data were calibrated in annual growth rates, in 5-year (1968–72, ... , 1993–97) partitions, for maximum likelihood estimation using flexible Box–Cox transformations model and bias-reducing jackknife resampling plan for data expansion. The drivers of OECD health care spending growth are economic and institutional. Findings from the growth convergence theory affirm that health care expenditure growth accords with conditional β convergence. Statistical significance and optimal functional form models are not unique across the growth period models. Our findings exemplify the benefits of jackknife resampling plan for short data series, and caution researchers against imposing faulty functional forms and applying asymptotic statistical methods to short time-series regressions. Policy implications are discussed.

Managed care, deficit financing, and aggregate health care expenditure in the United States: a cointegration analysis.

We applied a battery of cointegration tests comprising those of Johansen and Juselius [19], Phillips and Hansen [35], and Engle and Granger [6], to model aggregate health care expenditure using 1960–96 US data. The existence of a stable long-run economic relationship or cointegration is confirmed, in the United States, between aggregate health care expenditure and real GDP, population age distribution, managed care enrollment, number of practicing physicians, and government deficits. The evidence of cointegration among these variables, chosen on the theoretical basis of prior studies, implies that while they are individually non-stationary in levels, together they are highly correlated and move, in the long run to form an economic equilibrium relationship of US aggregate health care expenditure. More specifically, and for the first time in this line of inquiry, (i) managed care enrollment is found to be negatively associated with the level of health care spending, (ii) supply disinduced demand effects of physicians tend to moderate health expenditure, and (iii) government deficit financing is positively related to health care spending. The observed sign and magnitude of the income coefficient are consistent with health care being a luxury good.

Health care expenditure inertia in the OECD countries: A heterogeneous analysis

Health care expenditure studies of the Organization for Economic Cooperation and Development (OECD) countries remain important because their findings often suggest cost containment and other policy initiatives. This paper focuses on the compatibility of OECD health data with the “expenditure inertia” (or lagged adjustments) hypothesis, by modeling individual country time‐series data of 21 nations for the 1960–1993 period. Maximum likelihood estimates of the Box–Cox transformation regression models reveal that: (a) the hypothesized impact of health “expenditure inertia” is both pervasive and strong, averaging 0.64 across the countries; (b) the real GDP elasticities of health care expenditures vary widely among the countries and average 0.34 in the short run – implying that health care is a necessity; (c) the long run GDP elasticities are less than 1 in 8 countries, unitary elastic in 8 countries and elastic in 5 countries – suggesting that health care is not universally a necessity or a luxury commodity for the OECD countries; (d) physician‐inducement effects (dis‐inducement in a few countries) are weak, with a mean elasticity estimate of 0.17; and (e) no unique functional form approximation model is globally compatible with the data across the countries. Health care cost containment policy implications of these findings are explored.

Cost-Output Relation, Technological Progress, and Clinical Activity Mix of US Hospital Pharmacies

The fairly recent discovery and growing use of costly, revolutionary biotechnology drugs (e.g., GM-CSFs) in hospitals represent a major technical innovation in clinical pharmacy intervention. This paper models the translog production cost impacts of ten distinct clinical pharmacy tasks including pharmacokinetics, consultations with MDs, monitoring drug interactions,..., and presents results on decomposed technical change, economies of scope, scale economies, total factor productivity, and the alternatively defined pairwise factor substitution possibilities. The model was fitted to 1981–1990 time-series/cross-sectional data of US hospital pharmacies (source: Eli-Lilly Co.®). Zellners ISURE (Iterative Seemingly Unrelated Regression Equations) system estimates reveal a non-homothetic underlying technology with biased and pure technical change effects dominating scale-augmentation. Certain clinical tasks increase costs but are cost-saving if jointly produced with selective others. One of the largest and statistically significant economies of scope estimates is between ‘monitoring drug interactions’ and ‘consultations with MDs’. Cost implications of findings are given in light of the recent trends in hospital pharmacy clinical activities, related professional health manpower, and piecemeal health policy reforms.

Logistic Regression and Probability of Business School Alumni Donations: Micro-data Evidence

This paper analyzes the propensity of business school alumni to give (or not give) cash donations to their alma mater. A utility maximization model is estimated using logistic regression and survey sample data of 1955–56 to 1990–91 graduates of a large US category I Carnegie Foundation research doctorate public university. Maximum likelihood estimates of the model parameters fitted the observed data well. The probability of alumni giving has positive and strong associations with: specific fields of major; time since alumni graduated; other family members graduating from this university; children who are 18 years or older not residing at home; number of other cash-giving alumni known; household income levels; occupations of alumni and spouses; giving of cash gifts to other educational institutions and a number of charitable organizations with global outreach; charitable volunteering of time; availability of matching gift programmes at work; and how alumni view their educational experiences at the business ...

Are Factor Substitutions in HMO Industry Operations Cost Saving

Past research on the potentials for cost-saving scale and scope economies in multiproduct HMO operations in the United States are incomplete in their economics of the underlying technology structure. This article exploits the translog cost model estimates of all past studies of HMO production to infer the extent to which pairwise factor substitutions (e.g., administrative services vs. medical care resources, e.g., hospital days, physician services) suggest potential for cost savings in groups and independent practice associations. Given the industrys nonhomothetic production over a 20-year period, the conceptually valid Morishima elasticity measure at constant output reveals limited cost-saving potentials from factor interchange and input demands. These opportunities differ for groups and IPAs across Medicare and non-Medicare products. My findings add a timely and significant dimension to understanding potential cost savings in HMO operations. Policy suggestions and cost implications are rationalized in light of the declining Medicare HMO enrollment and recent changes in factor input prices.

Countercyclical union wage premium? Evidence for the 1980s

Empirical results based on pooled male data from the Panel Survey of Income Dynamics indicate an overall union wage premium of about 11.92 percent for the 1980s. In response to fluctuations in local labor market conditions, proxied by the local unemployment rate, a much more flexible wage-setting process is found in the nonunion sector relative to the union sector. The long-term effect of unemployment on nonunion real wages suggests an approximate 0.6 percent decline for every one percentage point increase in unemployment, a statistically significant reduction, but the long-term effect of unemployment on real wages of union members is negligible. The union wage premium ranges between 11.6 to 12.3 percent for the sample years. Even though union wages are insensitive to short-run fluctuations in local labor market conditions, and are somewhat countercyclical in nature, widespread union wage concessions which occurred during the 1980s may now be exerting a downward pressure on union wages.