HMGT 495 WK 2 DISC 2

CHAPTER

199

13ASYMMETRIC INFORMATION AND INCENTIVES

Learning Objectives

After reading this chapter, students will be able to

• define asymmetric information and opportunism, • describe two strategies for aligning incentives, • explain why opportunism is a special management challenge in

healthcare, and • discuss challenges in limiting opportunism.

Key Concepts

• Asymmetric information is information known to one party in a transaction but not another.

• Asymmetric information allows the better informed party to act opportunistically.

• Asymmetric information is a common problem for managers. • Aligning incentives helps reduce the problems associated with

asymmetric information. • Concerns about risk, complexity, measurement, strategic responses, and

team production limit the extent of incentive-based payments. • Incentive-based contracts have become more common in healthcare.

13.1 Asymmetric Information

Asymmetric information confronts healthcare managers in most of their professional roles. Vendors typically know more about the strengths and weaknesses of their products than do purchasers. Employees typically know more about their health problems than do human resource or health plan managers. Subordinates typically know more about the effort they have put

asymmetric information Information known to one party in a transaction but not another.

Lee.indd 199 1/2/19 3:15 PM

C o p y r i g h t 2 0 1 9 . H e a l t h A d m i n i s t r a t i o n P r e s s .

A l l r i g h t s r e s e r v e d . M a y n o t b e r e p r o d u c e d i n a n y f o r m w i t h o u t p e r m i s s i o n f r o m t h e p u b l i s h e r , e x c e p t f a i r u s e s p e r m i t t e d u n d e r U . S . o r a p p l i c a b l e c o p y r i g h t l a w .

EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS AN: 2144510 ; Robert Lee.; Economics for Healthcare Managers, Fourth Edition Account: s4264928.main.eds

Economics for Healthcare Managers200

into their assignments than do their superiors. Providers typically know more about treatment options than do their patients. In all these examples, one party, commonly called an agent, has better information than another party, commonly called a principal. Unless the principal is careful, the agent may take advantage of this information asymmetry—in other words, engage in opportunism.

Asymmetric information can result in two types of problems. One is that mutually beneficial transactions may not take place if concern about asymmetric information is too great. The other is that resources may be wasted because of agents’ opportunism or principals’ costly precautions. For example, an insurer cannot easily discern whether a treatment is really needed (Arrow 1963). In response, an insurer may not cover services thought likely to be abused, may require substantial consumer payments to restrain demand, or may require prior authorization before providing coverage. As a result, consumers may not use helpful services because the services cost too much. Alternatively, the plan, providers, and consumers may experience increased costs due to the requirement for prior authorization. (The insurer must staff the authorization office, the provider must spend time and money getting authorizations, and the consumer is likely to experience delays and repeat visits.) Asymmetric information also affects managers directly. Man- agers are often poorly informed about the quality, efficiency, and customer satisfaction issues that their subordinates face. But managers are also often poorly informed about whether costs are padded, whether quality problems are avoidable, or whether staffing is adequate. Fearing that subordinates will take advantage of them, managers may require reviews or audits. Both increase costs without directly adding to the output of the organization.

Asymmetric information is a concern when

• the interests of the parties diverge in a meaningful way, • the parties have an important reason to strike a deal, and • determining whether the explicit or implicit terms of the deal have

been followed is difficult.

These circumstances are far from rare. Unfortunately, they are an invitation to act opportunistically.

13.2 Opportunism

Opportunism can take many forms. Crime is one. For example, deliberately billing a health plan for services that were not actually rendered is a form of opportunism more commonly known as fraud. The forms of opportunism

agent A person who provides services and recommendations to clients (who are called principals).

principal The organization or individual represented by an agent.

opportunism Taking advantage of a situation without regard for the interests of others.

Lee.indd 200 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 13: Asymmetr ic Information and Incentives 201

that managers deal with are not usually so stark. Cruising the Internet rather than making collection calls, using the supplies budget to refurbish your office, scheduling a physical therapy visit of questionable value to meet vol- ume targets, and referring a patient to a specialist for a problem you could easily handle are also examples of opportunism.

From experience, we know that some individuals are opportunistic some of the time. Some individuals seldom act opportunistically, whereas others often do. As a first step, we try to avoid dealing with those who are the most opportunistic. We then try to set up systems to restrain those who may be tempted. These systems will be imperfect because our ability to anticipate what may happen and how individuals may react is imperfect.

13.2.1 Remedies for Asymmetric Information Remedies for asymmetric information focus on aligning the interests of the parties or monitoring the behavior of the agent. Changes in incentives are usually part of the preferred strategy because monitoring is usually expensive and nonproductive. For example, healthcare plans are commonly subject to utilization review designed to control use of services. Utilization review rarely changes recommended therapies, however, despite its cost and annoyance. Health plans would love to eliminate utilization review. Without it, a plan would rapidly gain market share because it could increase consumer satisfac- tion, increase provider satisfaction, and reduce premiums. In addition to being costly, monitoring may be difficult. For example, a product that a ven- dor honestly recommended may fail or may not meet your needs, or it may work but have features you do not need and cost more than a more suitable product. Monitoring is likely to be only part of the remedy for asymmetric information.

13.2.2 The Special Challenges for Healthcare The challenges posed by asymmetric information are not unique to health- care, although their extent poses special problems for healthcare managers. Three features make asymmetric information especially troublesome in the healthcare sector:

1. By paying the bills of healthcare providers, insurance creates a principal–agent relationship not found in most fields.

2. Insurance reduces the patient’s incentive to monitor the performance of healthcare providers because it limits the patient’s exposure to financial opportunism.

3. Asymmetric information is intrinsic to most provider–patient relationships. Patients typically seek providers’ services because they want information, so opportunism is always possible.

Lee.indd 201 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Economics for Healthcare Managers202

Opportunism is such an obvious risk that strategies have developed to limit it (Arrow 1963). One of the most obvious is our preference for dealing with those who have proven themselves. For example, primary care physicians tend to refer patients to physicians who have served them and their patients well. For fear of losing this business, specialists who might be tempted to pro- vide unnecessary services will be reluctant to do so. These sorts of ongoing relationships—between buyer and seller, patient and provider, and supervisor and subordinate—tend to deter observable opportunism. Much of the regu- lation of the healthcare sector also serves to deter opportunism. The problem is that these mechanisms work only when opportunism is detectable. In many cases, it is not.

13.2.3 Signaling When differences in quality or other attributes of care are hard to observe, agents may use signaling to reassure principals. Signals should tell prospec- tive clients about the agent, should be hard to counterfeit, and should be relatively inexpensive. Brand names are classic signals. Including a Pfizer label on a new drug costs little and reassures consumers that the drug meets stringent quality standards because substandard quality would hurt Pfizer’s sales. The challenge is to prevent others from counterfeiting the labels. Sur- prisingly, branding in the healthcare market, especially branding of healthcare services, is not common. Quality certification is another strategy for dealing with asymmetric information. For example, hospital accreditation by The Joint Commission is a signal of quality that is difficult to counterfeit. Unfor- tunately, the process is so expensive that many smaller hospitals do not seek accreditation.

Other signals may be useful but are likely to be less credible. For example, high prices and high levels of advertising also serve as quality sig- nals because low-cost, low-quality providers could not afford to advertise frequently or raise prices (Gneezy, Gneezy, and Lauga 2013). In markets with standardized products, poorly informed agents can buy information (e.g., by subscribing to Consumer Reports) or copy well-informed agents. The more individualized products are, the less this strategy works, so its value in the healthcare market is unclear. Although we can identify healthcare cases in which signaling reduces the problems associated with asymmetric informa- tion, it is far from a comprehensive solution.

13.3 Incentive Design for Providers

Recognition that the insurance system of the United States created multiple incentives for inefficiency triggered the growth of managed care. Providers

signaling Sending messages that reveal information another party does not observe.

Lee.indd 202 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 13: Asymmetr ic Information and Incentives 203

were faced with strong incentives to deliver care as long as the benefits exceeded their patients’ costs, and costly care was often free for insured patients. Neither party had a compelling reason for taking the true cost of care into account. We have already discussed redesign of consumer payments, so let us consider how incentives relate to provider payments.

Incentives are implicit in the four most common methods of paying providers: volume-based payments, salaries, capitation, and case-based pay- ment. Each of these methods has some advantages and disadvantages.

Exhibit 13.1 contrasts the incentives created by different payment sys- tems. Note that volume-based and salary compensation systems incorporate opposite incentives. The incentive structures of capitation and case-based payment systems are similar and fall between these opposite cases. Volume- based, case-based, and capitation payment systems immediately reward pro- viders who have large numbers of clients. Having more clients means higher revenues in all these systems. In contrast, unless other incentive systems are in place (e.g., review by superiors or the possibility of promotion), salary and budget payment systems do not reward providers according to the number of clients they serve.

The only form of payment that rewards providers who provide a large volume of services per client is volume-based payment. In case-based and capitation systems, the disincentive for high volumes of service per client is tempered by the rewards for attracting additional clients. Just as they do not reward for large numbers of clients, salary and budget systems also deter providers from delivering high volumes of service per client.

All the payment systems except volume-based payment encour- age providers to avoid clients with complicated, expensive problems (or at least encourage them to prefer clients with simple, inexpensive problems).

capitation Payment per person. (The payment does not depend on the amount or type of services provided.)

case-based payment A single payment for an episode of care, regardless of the number of services.

Volume- Based

Case- Based Capitation

Salary or Budget

Number of clients + + + –

Services per client + – – –

Client acuitya + – – –

Unbillable servicesb – + + +

Note: A “+” indicates that the compensation system rewards producing more of an output or using more of an input. A “–” indicates that the compensation system rewards producing less of an output or using less of an input. a In this context, client acuity refers to the amount of services that a client is likely to need. Higher acuity means that a client is likely to need more services. b Unbillable services include both services for which the provider cannot bill because of the provisions of the insurance plan and services provided by others.

EXHIBIT 13.1 Financial Incentives of Alternative Compensation Systems

Lee.indd 203 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Economics for Healthcare Managers204

Expensive clients, when combined with fixed payments per case or per period, are financially unrewarding for providers. Likewise, all the payment systems except volume-based payment motivate providers to refer patients to external services (e.g., church-sponsored organizations or services provided by friends), as long as they are cost-effective from the provider’s perspective. From society’s perspective, patients should be referred elsewhere as long as the marginal benefit of doing so exceeds the marginal cost. Providers who are paid on the basis of cases, capitation, or salary may refer patients too often, especially if the provider does not bear the full cost of the services of community organizations or other external services. In contrast, only billable services can be profitable in volume-based systems. Volume-based payment creates an incentive not to use external resources (or at least not to use the organization’s resources to improve clients’ access to them). Volume-based payment typically rewards providers who refer patients too infrequently, from society’s perspective.

None of these payment systems solves the asymmetric information problem. Providers still usually know more about appropriate treatment options than do patients or insurers. Volume-based providers inclined toward opportunism are still able to recommend additional billable services, case- based providers are still able to avoid unprofitable cases, capitated providers are still able to recommend limited treatment plans, and salaried providers are still able to limit how much they do.

This discussion should not be construed as an assertion that only financial incentives matter. Such an assertion would be inconsistent with basic economic theory, which postulates that principals and agents balance alterna- tive objectives. Only some of these goals will be financial. For example, some physicians may offer extensive patient education programs because of their commitment to the health of their patients or because the programs are an effective marketing tool, even if the volume-based payment system does not treat these programs as a billable service. Nonetheless, economics anticipates an aggregate response to financial incentives and predicts that physicians will offer more of such services if the volume-based payment system offers com- pensation for them or if they are profitable under case-based or capitation arrangements.

Suppose that a physician schedules four patients per hour for 30 hours per week and works 48 weeks per year (see exhibit 13.2). Under plan A, the physician earns $20 per patient and has a total income of $115,200. (This example bases compensation on visits to simplify the discussion, not to define an attractive volume-based compensation plan. More sensible volume-based payment systems base compensation on billings, relative value units, and so forth.) Plan B provides a base salary of $80,000 plus $20 per patient for visits in excess of 4,000. At the margin, plans A and B have the same incentives,

Lee.indd 204 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 13: Asymmetr ic Information and Incentives 205

even though plan B combines salary- and volume-based payments. Each plan pays $20 per patient and provides the same total income. This example illus- trates that blended compensation systems can give agents similar incentives with less risk than pure compensation systems.

The incentives of plan C are subtly different from the incentives of plans A and B. Plan C offers a $100,000 base salary plus $20 per patient for visits in excess of 4,000. Although plan C pays $20 per visit at the margin, as plans A and B do, the physician’s income will be higher under plan C than it would be with the same number of patients under plan A or B. Consequently, the physician may feel less need to add an additional patient at the end of the day or double book to squeeze in an acutely ill patient. In this case, income effects are the effects incentive systems have on physicians’ decisions about the number of patients they will treat (Kalb et al. 2017).

Plan D offers a base salary of $57,600 plus $10 for each patient visit. Even though the physician’s income will be the same with 5,760 patients per year under plans A, B, and D, the physician may choose to see fewer patients under plan D because the marginal reward is smaller.

13.4 Insurance and Incentives

How much have compensation systems changed since 2010, given the importance of how providers are paid? Less than you might suspect, but change is taking place. For example, even in areas in which managed care is pervasive, most physicians continue to be paid on the basis of their productiv- ity (typically measured by billings, visits, or net revenue), just as they were at the turn of the century (Landon and Roberts 2013). A number of organiza- tions are changing their compensation models, but they are headed in differ- ent directions. Some are moving to salaries without quality or productivity bonuses. Some are moving to a mix of quality-based and productivity-based compensation (Barkholz 2017).

In 2015 Congress passed legislation called the Medicare Access and CHIP Reauthorization Act, commonly called MACRA. While still basing

income effect The effect of income shifts on amounts demanded or supplied. (Shifts may be due to changes in income or due to changes in purchasing power caused by price changes.)

Plan Base Salary Marginal

Compensation Volume

Payments Total Income

A $0 $20 $115,200 $115,200

B $80,000 $20 $35,200 $115,200

C $100,000 $20 $35,200 $135,200

D $57,600 $10 $57,600 $115,200

EXHIBIT 13.2 An Illustrative Model of Incentives

Lee.indd 205 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Economics for Healthcare Managers206

most Medicare payments on volume, MACRA adds significant incentives for quality. Physicians with high quality scores can get bonuses that start at 4 percent in 2019 and increase to 9 percent in 2022. Physicians with low qual- ity scores can get penalties of the same size (Mire and Desai 2017). There are two ways of earning bonuses. Physicians can report their performance using the Merit-based Incentive Payment System, or they can work in an organization that uses an alternative payment model (typically an ACO) and get a 5 percent bonus each year because their performance is monitored by the ACO. Currently, the incentives are modest, and the effects are likely to be modest as well. But, as case 13.1 suggests, both the incentives and the effects may grow substantially.

Incentives in Accountable Care Organizations

An accountable care organization (ACO) represents a consortium of doctors, hospitals, and other providers who contract to take financial responsibility for the quality of care. As such, it faces a complex incen- tive problem, needing to identify contracts that allow it to meet its quality and cost targets. An ACO also needs to establish contracts with providers that incentivize them to provide efficient, high-quality care. In addition, the ACO must do this as the providers are simultaneously being paid using traditional volume-based methods, bundled pay- ments, and a variety of alternative payment models. Whitman (2017) quotes Micky Tripathi, founder and CEO of the Massachusetts eHealth Collaborative, as saying, “What makes a successful ACO? As an indus- try, we don’t know.”

A consortium that decides to form an ACO contracts with Medicare for three years. Medicare sets a benchmark that is a weighted average of spending by beneficiaries that are attributed to the ACO. One impli- cation is that consortia with high levels of spending in the past find it easier to meet cost goals. In contrast, commercial ACO contracts are based on a negotiated rate and a negotiated degree of risk.

For example, UnityPoint Health, a $4 billion system with 43 hospi- tals across Iowa, Illinois, Wisconsin, and Missouri, has a physician-led ACO that in 2016 covered 70,672 Medicare beneficiaries in the Medi- care Next Generation ACO model. Through its self-insured health plan

Case 13.1

(continued)

Lee.indd 206 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 13: Asymmetr ic Information and Incentives 207

and its ACO contracts with private payers, includ- ing UnitedHealthcare, Wellmark, and Blue Cross and Blue Shield of Illinois, it covered another

255,379 individuals, with 102,125 of those in full-risk contracts (Uni- tyPoint Accountable Care 2017). UnityPoint did not reduce costs in its first attempts at a Medicare ACO but ultimately developed a system for reducing costs. The system tracks high-risk patients, standardizes care pathways, uses analytics to understand the population it serves (and the care being provided), and emphasizes incorporating behavioral health into primary care.

UnityPoint Clinic, which has 500 physicians across Iowa and Illi- nois, now bases 86 percent of compensation on productivity. By 2020 it plans to have 33 percent of physician compensation based on pro- ductivity, 33 percent based on salary, and 33 percent based on cost, quality, and satisfaction measures (Barkholz 2017). The logic is that provider payments need to be aligned with payments for care.

Discussion Questions • Why is creating a successful ACO difficult?

• How many Medicare ACOs are there? How are they structured?

• How many commercial ACOs are there? How are they structured?

• How many Medicaid ACOs are there? How are they structured?

• Is the number of ACOs increasing or decreasing?

• What outcomes represent success for an ACO? What predicts success?

• Do ACOs that accept more risk get more shared savings? Why?

• Do some ACOs improve clinical quality? How do they do it?

• Do some ACOs improve patient satisfaction? How do they do it?

• How does being paid in varied ways complicate ACO design?

• How does being paid in varied ways affect provider incentives?

• What sort of incentives does pure volume-based payment create for providers?

• What sort of incentives does a mixed payment model create for providers?

• What sort of incentives does pure salary create for providers?

• How does MACRA affect incentives to create a Medicare HMO?

• What incentives does the Medicare benchmark create?

Case 13.1 (continued)

Lee.indd 207 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Economics for Healthcare Managers208

13.5 Limits on Incentive-Based Payments

A number of factors limit how complete incentive-based payments can be. Concerns about risk, complexity, and team production make agents reluctant to enter into incentive-based compensation arrangements. Likewise, concerns about opportunism make principals reluctant because a high-powered incen- tive system may leave them worse off if agents respond in unanticipated ways.

13.5.1 Risk Capitation, utilization withholding, and case-based payment systems are often referred to as risk-sharing systems. This term is somewhat misleading. The goal of these systems is incentive alignment; risk sharing is a side effect. For exam- ple, capitation gives physicians incentives to use resources wisely, so capitation succeeds if physicians do not run unnecessary tests or if they avoid hospital- izing patients when better community treatment options are available. Full- risk capitation, in which physicians are responsible for all their patients’ costs, gives physicians incentives to take such steps. Unfortunately, the financial risks associated with full-risk capitation can be substantial. One patient with a rare, expensive illness can bankrupt a solo practice; an unexpected jump in pharma- ceutical prices can bankrupt a small provider-owned HMO. These risks are one reason the growth of capitation has stalled and many organizations avoid full- risk capitation (Mechanic and Zinner 2016). Organizations that accept full-risk capitation are organized differently from those that do not. They are generally larger, more likely to have salaried physicians, more likely to have sophisticated information management, and more likely to have programs to reduce treat- ment variation and manage high-risk patients (Mechanic and Zinner 2016).

13.5.2 Complexity Providers and employees are more likely to respond to simple, comprehensi- ble systems than to complex, confusing systems. Simple systems limit the use of incentives and the problems they create. If you want the payment system to reward physicians for keeping customer satisfaction high, MMR (measles, mumps, and rubella) vaccination rates high, out-of-formulary drug use low, hospitalization rates low, hospital lengths of stay short, after-hours response times prompt, record updates prompt, and asthma follow-up appointments timely, the system is likely to be unwieldy. Moreover, the reward associated with each component of the system is likely to be small.

13.5.3 Opportunistic Responses Managers must anticipate opportunistic responses to incentive systems. An agent with better information can harm the principal. In many cases, whether

Lee.indd 208 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 13: Asymmetr ic Information and Incentives 209

an agent has lived up to contract requirements is difficult to ascertain. In other cases, the agent may act in ways the principal did not anticipate. Some responses will necessitate system redesign; some will have to be tolerated to prevent the system from becoming excessively complex. For example, one response to the price reductions introduced by PPOs was to unbundle services. Physicians and other providers began to bill separately for services once included in the standard office visit. While insurers attempted to limit unbundling in a variety of ways, the fundamental problem remained that the incentives of physicians and insurers were misaligned (Golden, Edgman- Levitan, and Callahan 2017). Physicians’ profits would be higher when they billed for more services, but insurers’ profits would be higher when physi- cians billed for fewer services.

13.5.4 Team Production Team production also limits the use of incentives. Production of health- care products usually involves a number of people, and the shortcomings of one person can undermine the efforts of the entire team. For example, rudeness by one disaffected team member can negate the efforts of oth- ers to provide exemplary customer service. This interdependency can also weaken the effects of individual incentives. Workers who try hard to do a good job or physicians who are conscientious about reducing length of stay are likely to feel that their efforts are not appreciated if the shortcomings of others deny them bonuses. Building and maintaining effective teams are important tasks for managers. Unless carefully structured, financial incentives tend to reward individualistic behavior, which usually weakens teams. Equally problematic, team financial incentives (i.e., every member of the team receives a bonus when the team reaches its goals) often fail to motivate workers.

13.6 Incentive Design for Managers

Incentives for managers can be financial or nonfinancial. If both types are used, the two incentive systems should operate in tandem. Otherwise, they may worsen the problems created by asymmetric information (Lagarde and Blaauw 2017).

Incentive pay for managers is a partial response to the asymmetric information problem. It usually takes the form of bonus payments, profit sharing, or stock options. In most cases, it is a modest part of total com- pensation and is only loosely tied to managers’ performance. Four concepts underlie incentive pay for managers:

Lee.indd 209 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Economics for Healthcare Managers210

1. Financial incentives can strongly motivate people to perform in ways the organization desires, yet organizations seldom want managers to focus only on duties that will increase their pay. (Volume-based compensation presents the same problem.)

2. Managers’ goals are often not fully defined. Managers need to respond creatively to problems or, better yet, position the organization to respond to problems that are not yet evident. Performance assessment based on intangibles would be difficult, if not impossible.

3. Most managers’ performance is hard to measure. As a result, compensation based on individual productivity ceases to make sense.

4. What is measurable and what is desired are unlikely to coincide. Compensation based on measurable outputs is likely to increase opportunism as managers react to what is rewarded rather than to what is sought.

For these reasons, incentive pay for managers generally needs to reflect the success of the overall organization. The dilution of incentives that results from using profit sharing or gainsharing is a reasonable price to pay for promoting team-oriented behavior. Gainsharing is like profit sharing, but bonuses can be based on a broader array of outcomes. (Chapter 6 discussed gainsharing in not-for-profit hospitals, where profit sharing is not permitted.) Members of a group can earn bonuses for hitting production, customer satisfaction, profit, quality, or cost targets. As individual contributions become less discernible, group incentives are likely to become more effective. Members of the group will be able to monitor each other more easily, alignment of the group’s and the organization’s incentives will become more important, and the group will more easily alter how it does its work. For example, hospital care is produced by teams, but pay for many physicians depends on their personal billings. To encourage physicians to participate in hospital performance improvement activities, implementing payments to physicians that are based on the perfor- mance of the hospital is often helpful.

Incentive pay is only part of an effective incentive system. Economic theory does not imply that individuals will not respond to opportunities to do challenging work, public celebrations of their accomplishments, or a positive review by a trusted mentor. An effective manager will consider these tools as well. Successful organizations require cooperation in management and production, so a nonfinancial system that rewards cooperation is a sen- sible option for aligning incentives. Promotions typically combine financial and nonfinancial rewards.

gainsharing A general strategy of rewarding those who contribute to an organization’s success. (Profit sharing is one form of gainsharing. Rewards can be based on other criteria as well.)

Lee.indd 210 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 13: Asymmetr ic Information and Incentives 211

The Total Care and Cost Improvement Program

In 2011 CareFirst BlueCross BlueShield, a plan offering coverage in Maryland, the District of Columbia, and northern Virginia, implemented a patient-centered medical home (PCMH) program that included enhanced payments for primary care practices, financial incentives for primary care physicians to reduce spending, and care coordina- tion tools to support improved care. The model, called the Total Care and Cost Improvement Program, changed payments to primary care physicians, increasing fees by 12 percent to compensate for extra care coordination and population management activities. The program also established a one-sided system of shared savings, increasing fees for the following year if spending was below the target. For example, savings realized in 2012 would increase fees in 2013. The increases depended on the savings achieved by the panel, a quality score, panel size, and savings consistency over time. The increases were substan- tial, an average of 45 percent by 2013 (Afendulis et al. 2017).

How well did the program work? That is not clear. Afendulis and colleagues (2017) conclude that savings were small and that many physicians were not fully engaged with the program. Cuellar and col- leagues (2016) report savings of nearly 3 percent in 2013, largely driven by reductions in emergency department and hospital use. They conclude that “a PCMH model that does not require practices to make infrastructure investments and that rewards cost savings can reduce spending and utilization” (Cuellar et al. 2016, 1382). The differences appear to be driven by analytic decisions about whether to analyze use of services by patients who switched to PCMH practices during the first three years (Afendulis et al. 2017).

The differing interpretations may not matter. In September 2017 CareFirst issued a press release hailing “an historic slowing of overall medical cost growth” (CareFirst BlueCross BlueShield 2017). In 2016 CareFirst members seeing PCMH providers had hospital admission rates that were 10.4 percent lower and readmission rates that were 34.7 percent lower than those of patients receiving care from other providers. What changed? First, PCMHs had become a much more common approach to primary care by 2016. Second, in 2014 CareFirst got a grant from Medicare to extend the program to cover Medicare beneficiaries. This change increased its clinical and financial effects.

Case 13.2

(continued)

Lee.indd 211 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Economics for Healthcare Managers212

The grant also allowed CareFirst to train staff in the PCMH practices and to hire care coordinators, nurse case managers, consultants, pharmacy man-

agers, clinical pathway specialists, and analysts. Third, organizational change takes time, and three more years had elapsed.

Bleser and colleagues (2014) argue that becoming an effective PCMH requires three things:

1. strong desire for change and a belief in the need for change,

2. capacity to carry out performance improvements, and

3. detailed understanding of current and best-practice clinical protocols.

It is not clear that these elements were present during the early years of the Total Care and Cost Improvement Program. Afendulis and col- leagues (2017) report that physicians in the program were receptive to making quality improvements but were less interested in cost reduc- tions. In addition, most were not familiar with or interested in the shared savings incentives.

Discussion Questions • Why would physicians not be interested in cost reductions?

• How should the program have been framed to align the goals of CareFirst and physicians?

• How did including Medicare beneficiaries change incentives?

• The Medicare grant added infrastructure to the program. Was that important?

• Have other PCMH programs become more effective as they matured?

• What should have been done to improve physicians’ knowledge of current clinical protocols?

• How could one improve physicians’ knowledge of best-practice clinical protocols?

• Why were most physicians unwilling to change their practices to claim incentives?

• Most PCMH programs pay care coordination fees per patient per month for patients with chronic illnesses. How do the incentives differ from those of the CareFirst program?

• Would care coordination fees have been a better strategy than higher visit fees?

• Many of the practices were small. Do you expect that they could carry out performance improvement projects?

Case 13.2 (continued)

Lee.indd 212 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 13: Asymmetr ic Information and Incentives 213

13.7 Conclusion

Incentive restructuring is an imperfect response to the problem of asymmet- ric information, as are all responses to this problem. The rewards of incentive systems are usually based on results, not on what agents actually do, and agents can respond opportunistically to virtually any incentive system. The challenge is to align the incentives of all the individuals in a system with the interests of its stakeholders. Because good incentive systems must balance competing objectives, no magic formula exists. In addition, managers must anticipate that incentives may have multiple effects and that designing incen- tive systems and keeping them up to date will be expensive.

Major changes in payment systems are underway. Both public and private insurers are moving aggressively away from pure volume-based pay- ments. Most of the new systems incorporate quality measures in some way, but quality is multidimensional and hard to measure. In addition, measuring quality uses resources. O’Shea (2017) estimates that quality reporting costs more than $40,000 per physician per year, and many are concerned about the burdens imposed on providers. MACRA creates significant incentives for physicians to participate in advanced alternative payment models, which could reduce the reporting burden and allow revenues to rise. The problems that asymmetric information creates have not gone away, but the new pay- ment systems have created pressure for change.

Exercises

13.1 Describe some healthcare situations in which an agent has taken advantage of a principal. Then describe some healthcare transactions that have not taken place because of fears about asymmetric information.

13.2 Identify some ways that nursing homes can signal high quality to consumers. Which of these signals are most apt to be reliable?

13.3 Provide an example of costly monitoring in the healthcare workplace. Can you think of an employment contract that would allow a reduction in monitoring without a reduction in quality?

13.4 What are some strategies for reducing adverse selection in insurance markets? What sorts of problems do these solutions cause?

13.5 One physical therapist is paid $20 per session. Another is paid $400 per week plus $20 per session in excess of 20 sessions per week. A third is paid $400 per week, plus $200 per week for having all paperwork complete and filed within 48 hours, plus $20 per session in excess of 30 sessions per week. How do the therapists’ incentives

Lee.indd 213 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Economics for Healthcare Managers214

to produce sessions compare? How do their incentives to complete paperwork differ?

13.6 One physician is paid $100 per visit. Another is paid $2,500 per week plus $100 per session in excess of 20 sessions per week. A third is paid $2,000 per week plus $100 per session in excess of 20 sessions per week. The third physician is also paid a weekly bonus of $500 for being in the top quartile for management of common chronic diseases, appropriate antibiotic use, preventive counseling, screening tests, and appropriate prescribing in elderly patients. How do the physicians’ incentives compare?

13.7 The Federal Trade Commission requires that firms advertise truthfully. Why does this requirement promote competition? Would firms be better or worse off if the Federal Trade Commission adopted a “let the buyer beware” policy?

13.8 Your firm sells backup generators to hospitals and clinics. The generators are guaranteed to operate on demand for two years. Your data show that the generators run an average of 42 hours per year. Your firm offers an extended warranty that covers the next three years. Your data show that repairs are needed for 2 percent of units during this three-year period. When repairs are needed, the average cost is $4,000. You charge $400 for the extended warranty, and about 20 percent of your clients buy it. a. The extended warranty has been a consistent money loser. Claims

average $1,000 per customer. How could this situation happen, given the data presented here?

b. Would raising the premium to $1,000 solve this problem? c. What would you recommend that your company do to solve this

problem? 13.9 For the population as a whole, average healthcare spending is

$1,190 per year. Those with a family history of cancer (5 percent of the population) spend $20,000 on average, and those with no family history (95 percent of the population) spend $200. An insurer is offering first-dollar coverage for $1,200. a. You are not risk averse and have no family history of cancer. Do

you buy coverage? b. You are not risk averse and have a family history of cancer. Do

you buy coverage? c. If you were risk averse, how would your answers to the last two

questions change? d. What could an insurer do to prevent this sort of adverse

selection?

Lee.indd 214 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 13: Asymmetr ic Information and Incentives 215

e. What would be wrong with having everyone undergo a physical exam to qualify for coverage?

13.10 You want to hire a new laboratory technician. Excellent technicians generate $1,000 in value added each week. Adequate technicians generate $500 in value added each week. Half the graduates are excellent, and half are adequate. a. You cannot tell who is highly capable and who is adequate. You

are prepared to pay each technician the value added. What salary do you offer?

b. Who will accept this offer? c. Is there any way that excellent technicians could communicate

their productivity? d. Propose a compensation system that will attract both types of

technicians and pay no technicians more than their value added. 13.11 A new test identifies individuals with a genetic predisposition to

develop heart disease before age 70. People who are predisposed to heart disease cost twice as much to insure as those who are not. a. Can you make a case that a law prohibiting this test would be a

good idea? b. The test is not expensive. Would you prefer to skip the test and

buy insurance at a premium that covers everyone or take the test and buy insurance at a premium that covers your group?

13.12 Your hospital wants to buy practices to expand its primary care networks. You are aware that physicians who want to sell their practices differ. Some love to practice medicine and love seeing patients. They want to sell their practices to focus on patient care 50 hours per week. Some physicians love to play golf and want to provide patient care no more than 35 hours per week. Propose a compensation plan that will allow you to hire only physicians who love to practice.

13.13 Having access to the books and understanding local markets better than new owners, the owners of medical practices generally understand their finances better than prospective buyers do. What sorts of transactions tend to take place as a result of this information asymmetry? What sorts of transactions tend not to take place? What can buyers and sellers do to offset this information asymmetry?

13.14 You are considering acquiring a firm rumored to have developed an effective gene therapy for diabetes. The value of the firm depends on this therapy. If the therapy is effective, the firm is worth $100 per share; otherwise, the firm is worth no more than $20 per share. Your company’s management and marketing strengths should

Lee.indd 215 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Economics for Healthcare Managers216

increase the share price by at least 50 percent in either case. You must make an offer for the firm now, before the results of clinical trials are in. The current owner of the firm will sell for the right price. Make an offer for the firm. Explain why you think your offer makes sense.

References

Afendulis, C. C., L. A. Hatfield, B. E. Landon, J. Gruber, M. B. Landrum, R. E. Mechanic, D. E. Zinner, and M. E. Chernew. 2017. “Early Impact of Care- First’s Patient-Centered Medical Home with Strong Financial Incentives.” Health Affairs 36 (3): 468–75.

Arrow, K. J. 1963. “Uncertainty and the Welfare Economics of Medical Care.” Ameri- can Economic Review 53 (5): 941–73.

Barkholz, D. 2017. “Changing How Doctors Get Paid.” Modern Healthcare. Pub- lished March 11. www.modernhealthcare.com/article/20170311/MAGAZINE /303119983.

Bleser, W. K., M. Miller-Day, D. Naughton, P. L. Bricker, P. F. Cronholm, and R. A. Gabbay. 2014. “Strategies for Achieving Whole-Practice Engagement and Buy-In to the Patient-Centered Medical Home.” Annals of Family Medicine 12 (1): 37–45.

CareFirst BlueCross BlueShield. 2017. “Patient-Centered Medical Home and Sup- porting Programs Contribute to Historic Slowing of Health Care Cost Growth.” Published September 18. https://member.carefirst.com/members /news/media-news/2017/patient-centered-medical-home-and-supporting -programs-contribute-to-historic-slowing-of-health-care-cost-growth.page.

Cuellar, A., L. A. Helmchen, G. Gimm, J. Want, S. Burla, B. J. Kells, I. Kicinger, and L. M. Nichols. 2016. “The CareFirst Patient-Centered Medical Home Program: Cost and Utilization Effects in Its First Three Years.” Journal of General Internal Medicine 31 (11): 1382–88.

Gneezy, A., U. Gneezy, and D. O. Lauga. 2013. “A Reference-Dependent Model of the Price–Quality Heuristic.” Journal of Marketing Research 51 (2): 153–64.

Golden, W., S. Edgman-Levitan, and S. R. Callahan. 2017. “Changing How We Pay for Primary Care.” NEJM Catalyst. Published November 20. https:// catalyst.nejm.org/changing-primary-care-payment/.

Kalb, G., D. Kuehnle, A. Scott, T. C. Cheng, and S. H. Jeon. 2017. “What Factors Affect Physicians’ Labour Supply: Comparing Structural Discrete Choice and Reduced-Form Approaches.” Health Economics 27 (2): e101–e119.

Lagarde, M., and D. Blaauw. 2017. “Physicians’ Responses to Financial and Social Incentives: A Medically Framed Real Effort Experiment.” Social Science & Medicine 179: 147–59.

Lee.indd 216 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 13: Asymmetr ic Information and Incentives 217

Landon, B. E., and D. H. Roberts. 2013. “Reenvisioning Specialty Care and Pay- ment Under Global Payment Systems.” JAMA 310 (4): 371–72.

Mechanic, R. E., and D. E. Zinner. 2016. “Risk Contracting and Operational Capabilities in Large Medical Groups During National Healthcare Reform.” American Journal of Managed Care 22 (6): 441–46.

Mire, R. D., and R. Desai. 2017. “How Physician Compensation in the United States Works.” Medscape. Published October 18. www.medscape.com/view article/886453.

O’Shea, J. 2017. “As MACRA Implementation Proceeds, Changes Are Needed.” Health Affairs Blog. Published April 21. www.healthaffairs.org/do/10.1377 /hblog20170421.059725/full/.

UnityPoint Accountable Care. 2017. Positioned for Success: 2015–2016 Value-Based Care Report. Modified August. www.unitypoint.org/filesimages/Accountable %20Care%20Section/000498-4%20UPHP%20ACO%20Value%20Report %202016_V05F_web.pdf.

Whitman, E. 2017. “A Tale of Two Accountable Care Organizations.” Mod- ern Healthcare. Published February 4. www.modernhealthcare.com/article /20170204/MAGAZINE/302049983.

Lee.indd 217 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Lee.indd 218 1/2/19 3:15 PM

EBSCOhost - printed on 1/12/2023 10:25 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use