eco question

Health Economics ECON 5860 PROF. KURT LAVETTI

• Based on these 3 characteristics, consider a “stock of addictive capital” S, such that the more addicted you are, the larger S is

• Suppose that whenever S is higher, you want to consume more • Consumers maximize utility over an additive good C, other

consumption goods Y, and addictive capital S U(C(t),Y(t),S(t))

• The stock of addictive capital depreciates over time at the rate δ

S(t)-S(t-1)=C(t)-δS(t)

• The rational addiction model assumes that consumers are forward looking, and anticipate the effect of consumption on their addictive capital S.

Model of Rational Addiction

Equilibrium in the Rational Addiction Model

• Predictions from this model: • Consumption at any point in time is related to current prices and

past prices (because past prices determine current addictive stock)

• Current consumption depends on anticipated future prices

• Permanent price changes have larger effects than temporary ones

• Long-run price elasticity of demand > short-run price elasticity of demand • The size of the difference increases with addictiveness

Rational Addiction Model

• Empirical test: do consumers respond to tax increases that have been enacted, but are not yet in effect?

• Yes: consumers purchase more cigarettes before the tax increase, but consume fewer cigarettes

• Increase in purchases consistent with stockpiling to avoid taxes

• Decrease in consumption is consistent with model of rational addiction

Testing the Rational Addiction Model

Policy Interventions: Excise Tax

Effect of excise tax on consumption depends on the relative size of the elasticity of demand to the elasticity of supply

Policy Interventions: Excise Tax

Source: Cawley and Ruhm (2012)

Policy Interventions: Excise Tax

Source: Cawley and Ruhm (2012)

• Practice Questions: • What would happen in the model if an addictive substance

becomes “more addictive”?

• What happens if the rate of depreciation of addictive capital decreases (so that it takes longer for addictive cravings to dissipate)?

• What might happen if two different addictive goods are partial substitutes (eg. prescription opioids and heroin) and the price of one good increases?

Rational Addiction Model

THE SUPPLY SIDE OF HEALTHCARE MARKETS

PHYSICIAN LABOR MARKETS

Physician Labor Markets

 Common view that there is a “shortage” of doctors and nurses in the US

 What is a shortage?  Shortage is a perception that the quantity outcome in the

market is lower than it “should be”

 How can you tell if there’s a shortage?  Several possible indications could be that a patient has to

wait “too long” for an appointment to see a doctor  May have to drive “too far” to see a doctor

12

Geographic Disparities 13

Entire county is Primary Care Shortage Area (PCSA) Part of county is a PCSA Not a PCSA

Center for Medicare and Medicaid Services (CMS) has Provider Incentive Payments to encourage doctors to move to shortage areas

Physician Shortages in the Media 14

Source: http://healthsciences.utah.edu/notes/images/transparency. pdf

Labor Shortages 15

S

D=MRP

q

W

W1

q1S q1D

shortage

• A shortage means that at a given price, the quantity supplied doesn’t equal the quantity demanded

Labor Shortages

 Shortages cannot exist in the long run in an efficient market with free entry

16

S

D=MRP

W2

q

W

q2

W1

q1S q1D

shortage

Adjust wages

Dynamics of Supply Shortages

 In the short-run shortages could occur if it takes time for supply to react to a change in demand  Training physicians and nurses takes a long time

17

D1

q

W

q1

S1

W1

Dynamics of Supply Shortages

 Suppose there is an unexpected shock to demand  In the short-run supply can’t respond  There must also be something restricting wages from rising to

w2, creating a short-run shortage

18

D1

q

W

W2

q1 q2

S1

W1

D2

“Shortage” caused by combination of: 1. Unexpected demand shock 2. Inability for supply to respond

quickly 3. Something preventing

market prices from adjusting to competitive equilibrium price

One Policy Solution: Increase Wages 19

• Why are there “too few” doctors? • Is it really because wages are too low? • US graduates plateau, but fraction of new MDs that are foreign-

trained doubled (21% to 45%) during the 1990s

Source: census.gov

One Policy Solution: Increase Wages 20

• Are wages really the problem though? • If there is a “shortage” in orthopedists, should the government intervene

to increase salaries above $413,000 per year to incentivize more orthopedics students?

Source: Medscape Compensation Report 2014

One Policy Solution: Increase Wages 21

• An international perspective on physician salaries in the US

Policy Question: 22

Entire county is Primary Care Shortage Area (PCSA)

Part of county is a PCSA

Not a PCSA

Suppose we want to set policies to fix the geographic disparities in physician supply • What is an

example of a policy that could achieve this goal?

• How should we measure shortage areas?

• What kinds of statistics should we use?

The Trouble with “Shortages” 23

Source: Scheffler 2008

• Wait times (number of days required to wait in advance of appointment) are often highest in areas where there are more physicians

• In Boston the wait times to see a doctor are twice as long as the national average

• Should policymakers intervene to fix this shortage?

• What if you then find out that Boston has 7 times more cardiologists per person than Detroit, and 5 times more than San Diego. Is there still a “shortage”.

Alternative Policy Tool: Recruit Foreign Medical Graduates

 Important providers of care in the U.S.  Nursing

 Physicians (particularly primary care)

 Typically must pass U.S. exams  Ethical/development questions

24

25

26Non-Physician Healthcare Labor

• The supply of nurses and physician assistants relative to the number of doctors has remained fairly constant

• Average salaries of nurses and assistants have changed similarly to physician salaries

• Nurses earn about 33% of physicians, PAs earn about 37%

• Some states have begun allowing nurses and PAs to treat patients and prescribe drugs in effort to reduce shortages

• Stange (2012) finds that allowing more autonomy for NPs and PAs has no effect on utilization, access, prices, or use of preventative care

To Understand Shortages, Start at the Beginning: The Training of Physicians

Medical school

 Entry into med school is competitive and selective worldwide

 In the US, on average only 50% of applicants are accepted into at least one school

 Length of medical school varies across country

 US & Canada applicants must first get a bachelor’s degree

 European applicants go directly from high school

 Medical school can be very expensive

 US: Students pay about $140k -- $225k for four years

 Total cost of training a new physician, including residency, is around $500k - $1 million per physician

 European medical training often heavily subsidized

Residency

 In addition to classroom work, physicians-in-training must also gain hospital experience  Residency is a period of 3+ years on-the-job training

following medical school

 In order to specialize in a specific type of medicine, often must complete multiple residencies requiring a total of 3-10 years (in addition to 4 years of medical school)

Physician Earnings, and the Rate of Return to Specializing 30

• Although completing training in a specialty area takes a long time, is this the only reason why earnings are so much higher among specialists?

Source: Medscape Compensation Report 2014

Explanations for High Specialist Salaries

 Market explanations:  Longer hours

 More time in residency, which requires some compensating wage differential

 Differences in ability (dexterity, IQ, etc.)

 Barriers to entry that prevent competition

 Bhattacharya (2005) estimates the rate of return to specializing relative to becoming a primary care (“family practice”) physician

31

Bhattacharya (2005): Specialty Matching and “skill premiums”  Differences in skills (board exam test scores, medical school

performance) cannot explain differences in wages  One exception is for very specific subspecialties—these physicians earn

more in large part because only the best students can be admitted to these specialty programs

32

• Example: Row 1 says that for an average FP physician, if they were to have become a surgeon instead they would earn 2.74% more than the average surgeon based on observed skill differences

Bhattacharya (2005): Explaining Higher Salaries of Specialists 33

Adding in market explanations

Conclusion: Market explanations can explain less than 50% of earnings differentials

Hours of work

Observed differences

Unobserved characteristics that affect specialty decision

Adjusting for timing differences (longer training)

Bhattacharya (2005): Remaining Explanations for Earnings Premium

 If supply and demand factors only explain half of the earnings differences between primary care and specialists, what could explain the rest?

 2 main possibilities:  People dislike some aspect of specialist jobs (seems unlikely)

 Barriers to entry into specialties cause earnings to remain above the level that would occur in a competitive market with free entry

34

Barriers to entry

Barriers to entry may explain the high IRR  In 19th century, becoming a doctor was simple

 Anyone could do it, no regulation about training

 American Medical Association (1847)  Pre-req’s for medical school  4 years medical school  Require doctors to have a license to practice  1910 Flexner Report helped shut down low-quality

med schools

 Result: less med schools and less med students

More barriers to entry

 Caps on medical school class size  Doctors need license to practice on their own

 International med graduates  Long and arduous process to practice in the US

 Nurses and Physician Assistants  Limited in scope of practice

 Alternative medicine  Chiropractors, acupuncturists, etc. need licensure too

Effect of Supply Constraint

 If the AMA constrains quantity, we could see what looks like a persistent shortage

37q

W

WFixed

Q_fixed Q_Competitive

S

WCompetitive

D Shortage

Tradeoffs from barriers to entry  Because of barriers to entry, consumers have to

pay above the competitive price  Physicians therefore earn monopoly rents

 Def. wages above the competitive price due to artificial constraint of the market

 Barriers to entry ensure that physicians are qualified

PHYSICIAN BEHAVIOR

Outline

 Simple theory of physician choices  Supplier-induced demand hypothesis  Small Area Variations (SAV)

 Why are costs so high in McAllen, TX?

40

Some Basic Theory

 What do physicians maximize?  Physicians are both workers who supply labor and

often small business owners  As workers, physicians get utility from income and

leisure time  However, physicians also have the ability to

induce their patients to demand care, but dislike doing so

 Utility = U(Y, L, I)  Y is net income; UY >0 (where UY is ΔU/ΔY)  L is leisure; time spent not working; UL>0  I is the degree of ‘artificial demand creation’; UI < 0

41

Supplier-Induced Demand

 Relies on the familiar concept of information asymmetry between provider and patient

 Physicians can alter quantity demanded, all else equal, by varying the ‘accuracy of advice’  Empirical question is do physicians actually do this

 Supplier-induced demand hypothesis was developed to try to explain some strange empirical phenomena in healthcare markets, but each of these phenomena could have other explanations as well

42

Income vs. Inducement

 Trade off the utility gains from additional income earned from extra services with the disutility or conscience cost of inducing demand

 A decrease in the profit rate will, in theory, lead to more demand inducement

43

• m is profit rate per unit of care provided

• Q0 is quantity of care without inducement

• I is amount of care induced

• If m falls from m to m’ inducement may increase from IE1 to IE2

Question

How would you empirically test whether physicians induce demand?

44

Circumstantial evidence for induced demand and why it isn’t conclusive

Observation 1. Whenever hospitals expand

capacity, they are immediately full

2. More surgeons leads to more surgery

3. More doctors in town is correlated with higher prices

4. Doctors who own their own labs do more lab tests

Criticism There could have been excess

demand to begin with

Could be due to reverse causation

Could be due to geographic differences in quality or changes in quality over time

Could be reverse causation, also patients could value the convenience

Empirical Versions of “Demand Induction”

1. QD increases when provider supply increases, all else equal 2. Equilibrium price increases when provider supply increases

46

D1

D2

S1

S2

P

Q

Supply and Demand Model of Supply-Induced Demand

Empirical Evidence of SID

 Physician supply increases quantity of services  Fuchs: estimates the elasticity of surgical operations

per capita to surgeons per capita is 0.28

 That is, a 10% increase in the supply of surgeons increases the number of surgeries by 2.8%, after controlling for the fact that surgeons prefer to locate in areas with higher demand for surgery

 One theoretical explanation for this is the target income hypothesis  Physicians expect to make a certain income (after

large investment in training) and if their income is too low they induce demand to increase income

48

Supply-Induced Pregnancies?

 Dranove and Wehner (1994)  Use same statistical approach from other studies,

but test whether increase in supply of obstetricians can “induce demand”

 Find evidence that obstetricians do induced demand

 This is obviously not possible, so suggests that there is something wrong with the statistical model that is probably also wrong in the other studies as well

49

A Clever Test of Induced Demand: What Happens when Physicians become Patients?

 Johnson and Rehavi (2014) study the choice between caesarian section or traditional delivery during childbirth

 Caesarean sections are generally much more expensive and more profitable for the hospital and physician

 Do physicians overuse c-sections because they are more profitable?

50

Induced Demand for Profitable Services

 To study this question, look at choices made by female physicians (or just obstetricians) when they are giving birth themselves

51