You are on page 1of 11

# The Production, Cost, and Technology of Health Care Ch 6

## Dr. Katie Sauer Health Economics

Recall from chapter 5 that health is the output goal. Much attention is paid to the intermediate output, health care. Outline: I. Degree of Substitutability in the Production Process II. Costs III. Efficiency IV. Technology __________________________________________________________ I. Degree of Substitutability in the Production Process Health practitioners often argue that there is basically one way to treat a given illness. (monotechnic view). Economists often note that there is more than one way to skin a cat. Suppose that when treating one case, it is not possible to substitute any nurse hours for any physician hours: Inputs are needed in a fixed combination of p1/n1. Suppose that there is some degree of substitutability for nurse hours and physician hours when treating one case: The rate of substitution changes as we move along the curve. Even though substitution is possible, nurses and doctors are not equivalent.

What degree of substitution is possible? The work done by economists suggests that substitution possibilities could be substantial. Depending on the number of physician hours employed, one physician extender could substitute for 25 percent to more than 50 percent of a physicians services. (Brown,1988; Deb and Holmes, 1998; Liang and Ogur, 1987; Okunade and Suraratdecha, 1998)

## Dr. Katie Sauer Health Economics

- look at elasticity of substitution (s ) - measures the responsiveness of a (cost-minimizing) firm to changes in relative input prices %FactorInpu tRatio
s =
%Factor Pr iceRatio

The larger the value of , the greater the substitutability. If a firm is a cost-minimizer, then it would respond to changes in input prices. It would tend to shift from the costlier input to the relatively cheaper one. Example: - Suppose in order to treat 75 cases, a hospital employs 100 nurses and 100 physicians. -The annual salary for a physician is \$200,000 and is \$40,000 for a nurse. - Suppose that the elasticity of substitution is 0.6. If each physicians salary increases by \$20,000, how will the hospital respond? Intuitively we know that this hospital will substitute away from physicians and use more nurses. To find the new number of nurses and doctors, we need to calculate the %Factor Input Ratio. We know = 0.60. We can calculate the % Factor Price Ratio because we have salary data. initial factor price ratio = new factor price ratio = % factor price ratio =

## Dr. Katie Sauer Health Economics

The initial factor input ratio = The new factor input ratio = So the ratio of physicians to nurses now = ____. If the hospital uses one less doctor, what does that mean? So, in response to the increase in physicians salaries, the hospital can use ___ nurses to substitute for one physician. Change in cost to the hospital from the substitution: save \$220,000 from not paying 1 physician costs (\$40,000 )(___) = \$__________ for the new nurses net savings of __________ Empirical Evidence for Input Substitution in Hospitals

II. Costs production function: relates inputs to output cost function: relates costs to output Theoretically speaking, the cost function is only valid for cost-minimizing firms. Do non-profits minimize costs? Recall that the firms problem is to produce a certain level of output at the minimum cost.

## Dr. Katie Sauer Health Economics

The expansion path C-F-G shows the costminimizing combinations of capital and labor that can be used to produce 100, 150 and 200 physician office visits, respectively.

Suppose we are given the input prices r = \$1,200 and w = \$1,000 The cost of producing 100, 150 and 200 physician visits would be: 100: 150: 200:

This total cost curve leads to an average cost curve like this one. Economies of Scale: exist over the range of output where a firms long run average costs are falling (as a firm gets larger / produces more output, its average costs fall)

## Dr. Katie Sauer Health Economics

Economies of Scope: exist when the joint production of two goods costs less than producing each good separately - only possible for multi-product firms ex: pediatric care and geriatric care Qp = 100 Qg = 150 If the following is true, then economies of scope exist: TC( Qp=100, Qg=150) < TC(Qp=100, Qg=0) + TC(Qp=0, Qg=150) Issues in Hospital Cost Studies a. Long Run vs Short Run - clear in theory, what about in practice? b. Structural vs Behavioral cost functions - Structural functions are derived from economic theory. (isoquant, isocost) - Behavioral functions are an analysis of patterns in actual cost data. c. difficulties in all hospital studies - What do hospitals produce? (face very different case mixes) - How to treat quality? - Lacking in reliable measures of hospital input prices. - Physician input prices are omitted. Research Findings: The most recent research supports claims that economies of scale exist in hospitals. Preya and Pink (2006) studied costs of Canadian hospitals prior to a massive consolidation, finding large scale unexploited gains to strategic consolidation in the hospital sector. Dranove and Lindrooth (2003) studied a large number of hospital consolidations, they found significant, robust, and persistent savings for mergers, 2, 3, and 4 years after consolidation.

III. Efficiency Technical Inefficiency: a producer is not achieving the maximum output for a given input combination.

## Dr. Katie Sauer Health Economics

- production within a certain firm Allocative Efficiency: requires the efficient allocation of inputs between firms, and between outputs. Economic theory says that allocative efficiency will result when: - inputs are purchased in competitive markets - firms are minimizing production cost Each firm must respond optimally to changes in input prices.

Empirical Techniques for evaluating efficiency: Frontier: actual output or costs are compared to the best possible situation

## Dr. Katie Sauer Health Economics

Non-Frontier: actual output or costs for two or more groups of firms are compared, controlling for other variables We will focus only on frontier. Frontier: Data Envelopment Analysis (DEA) - the frontier for production is initially unknown - it is revealed as firms are observed - linear programming is used to construct the outer shell of data points

## Technical inefficiency is measured as the distance from the frontier.

Notable drawback: assumes all firms that are not on the frontier are inefficient Notable feature: do not need to make any assumptions about the underlying distribution of inefficiency non-parametric Frontier: Stochastic Frontier Analysis (SFA) - each firm is treated uniquely - assume each firm will be affected by a shock to its ability to provide care - when a firm is randomly shocked (to alter its production and cost), its best possible practice (frontier) is randomly shifted

## Dr. Katie Sauer Health Economics

Notable drawback: need to guess in advance the distribution of inefficiency Notable feature: treats each firm uniquely DEA and SFA are often used as complementary tools. Results of Hospital Efficiency Studies: Reported overall efficiency has been quite high. The earliest DEA study (Valdmanis, 1990) reported technical efficiency levels of about 90 percent, while Magnussons DEA (1996) study reached similarly high levels. SFA studies have tended toward similar levels; early SFA studies (Zuckerman, Hadley, and Iezzoni, 1994; and Folland and Hofler, 2001) found the sum of technical and allocative inefficiency to be only a little more than 10 percent. For-Profit vs Non-Profit Hospitals In nearly all recent studies, nonprofit and for-profit hospitals appear approximately equal in efficiency. While the earliest studies (Valdmanis, 1990; and Ozcan et al., 1992) found differences between samples of public and for-profit hospitals, studies since then found no significant differences (Sloan et al., 2001).

## Dr. Katie Sauer Health Economics

Burgess and Wilson (1998) concluded: We find no evidence that differences in ownership affect technical efficiency after controlling for other factors. Inefficiency and Quality Deily and McKay (2006) explain that hospital inefficiency may reduce the quality of care. - inefficiency measure was a highly significant and positive contributor to a measure of hospital mortality rates. Laine and colleagues (2005) attempted similar tests for long-term care. - no inefficiency effect on clinical quality, - inefficiency contributes to the prevalence of pressure ulcers, indicating poor quality of care was associated with technical inefficiency. IV. Technology A. Technology and Costs Does technological change increase or decrease costs? - if productivity is raised, then costs are lower - if quality is improved, then costs can be higher Holding quality constant, technological change means it is less costly to produce a given level of output. However, technology may change the mix of products or services available, which may increase or decrease the cost of treating a patient. Measuring the cost of treatment can be difficult. - treatments change over time - effectiveness of treatment changes - quality changes

## Dr. Katie Sauer Health Economics

B. Who adopts new technology and why? 1. Profit principle: adopt if revenues are expected to increase 2. Information channels: emphasizes the role of friends/colleagues/ journals/etc in decision making 3. Theory of innovation: adopt when PV of future profits is greater than zero - wait too long, miss out - wait somewhat, gain from others experience 4. Managed care: flattening of incentives can decrease interest in switching technologies Classic Pattern of Adoption (logistic function)

## Dr. Katie Sauer Health Economics

Summary: The production function, which summarizes the relationship of inputs and outputs, also embodies the technology. Technology that permits substitution between inputs provides better flexibility to the manager. Cost estimation describes the cost curves, which identify the economies of scale and scope. Health firms may differ in technology because the adoption of new technologies differs among firms and is never instantaneous.