eco question
shazuanzhe
Lecture12_HealthSpending.pdf
Health Economics ECON 5860
PROF. KURT LAVETTI
HEALTH SPENDING
Data on Spending
u Measuring health care costs and expenditures
u Current levels
u Growth Components
u Reasons for health care cost growth
National Health Expenditures as a Percentage of Gross Domestic Product, 1980 – 2017
Source: Centers for Medicare & Medicaid Services, Office of the Actuary.
9. 1% 9. 4% 10
.2 %
10 .3
% 10
.2 %
10 .4
% 10
.6 %
10 .8
% 11
.2 %
11 .6
% 12
.3 %
13 .0
% 13
.4 %
13 .7
% 13
.6 %
13 .7
% 13
.7 %
13 .6
% 13
.6 %
13 .7
% 13
.4 %
14 .1
% 14
.9 %
15 .4
% 15
.5 %
15 .5
% 15
.6 %
15 .9
% 16
.4 %
17 .4
% 17
.4 %
17 .3
% 17
.2 %
17 .2
% 17
.3 %
17 .6
% 18
.0 %
17 .9
%
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
20%
80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12 14 16
Pe rc
en ta
ge o
f G D
P
International Spending Comparison
Source: Kaiser Family Foundation
International Spending Comparison
Source: PBS
Rates of Growth
u Growth rates: More useful metrics
u Persistent meaningful reduction in spending would require a decrease in rate of growth
u Many policies that aim to reduce spending have instead resulted in one-time reductions in spending, followed by the same high growth rate
Growth is the Difference …
Source: Kaiser Family Foundation
NHE Components: 1970-2017
Select Spending Categories
1970 1985 2005 2017
$B %NHE
$B %NHE
$B %NHE
$B %NHE
Hospitals $27.6 $165.4 $611.6 $1,142.6 36.8% 37.6% 30.8% 32.7%
Physicians $14.0 $89.8 $413.0 $694.3 18.7% 20.4% 21.2% 19.9%
Pharmaceuticals $5.5 $21.1 $205.2 $333.4 7.3% 4.8% 10.1% 9.5%
Administrative $2.8 $25.6 $143
Not Available 3.7% 5.8% 7.2%
Home Health / Nursing Homes
$4.3 $37.3 $169.3 $263.3 5.7% 8.5% 8.5% 7.5%
Total NHE $74.9 $439.9 $1,987.7 $3,492.0
General Trends: Spending Components
u Hospitals share is slightly declining over time
u Outpatient care and physician services slightly increasing
u Pharmaceutical share drops and then increases strongly through 1990s
u Medicare Part D implemented in 2006
u Administrative share is increasing
u Main lesson: spending levels have increased rapidly in every category
NHE Spending Sources
u Private Funds
u Out-of-Pocket (OOP) Expenditures u Private Health Insurance
u Payments from insurers (Non-Profit and For-Profit), includes premiums
u Public Funds
u Federal u Medicare (Elderly / Disabled)
u Medicaid (“Matching” Payments to state programs for “poor”)
u Other Federal (Includes SCHIP)
u State and Local u Medicaid
u Other State and Local (e.g. local public health facilities)
Who pays for the nation’s health care costs?
Who really pays the bill? Financing Source of Expenditures
Funding Mechanism Ultimate Payer
Medicaid/ SCHIP
Federal and State Taxation
Taxpayers
Medicare Payroll tax General taxes Premiums
Taxpayers (Intergenerational Transfer) Beneficiaries
Private Insurance
Premiums Employees, Individuals (in case of individual market)
OOP Patient Individuals
Sources of Medical Spending: 1970-2014
Source: CMS (http://www.cms.hhs.gov/NationalHealthExpendData), Kaiser Family Foundation, AHA, US Census Bureau, Income, Poverty, and Health Insurance Coverage in the United States: 2005.
1970 1985 2005 2014 NHE $B $74.9 $439.9 $1,987.7 $3,000.0 (% GDP) 7.2% 10.4% 16.0% 17.5% Public Exp. $28.1 $177.4 $902.7 $1,350.0 (% NHE) 37.5% 40.3% 45.4% 45.0% Medicare $7.7 $71.4 $342.0 $618.7 (% NHE) 10.3% 16.2% 17.2% 20.0% Medicaid $5.2 $40.9 $311.0 $480.0 (% NHE) 6.9% 9.3% 15.6% 15.0% Private HI $15.5 $131.0 $694.4 $991.0 (% NHE) 20.7% 29.8% 34.9% 33.0% Out of Pocket Costs $24.9
$95.7 $249.4 $329.8
(% NHE) 33.2% 21.7% 12.5% 12.0% NHE/Pop $369 $1,818 $6,693 $9,523
Population (Mill.) 203 242 297 319
General Trends since 1970: Sources of Finance
u Increasing shift from private to public finance
u Insurance has actually become much more generous over time!
u Consumers paid 33% of medical expenses out of pocket in 1970, but only 12% in 2013
u Both Medicaid and Medicare doubled their share of total health expenditures
If we care about understanding changes in total expenditures, then we need to understand these two components:
Total Spending = Price*Quantity
• What explains/predicts ΔP? • What explains/predicts ΔQ? • The difference between ΔP and ΔQ matters!
Reinhardt et al (2004): Why is US spending so high, and can we afford it?
u Ability and willingness to pay u High GDP per capita in US
u Distribution of market power and prices u Salaries of healthcare professional
u Fragmentation of purchasers
u Capacity of healthcare system u Supply side: “If you build it they will come”
u Administrative costs u Extremely complex and fragmented system
u Unwillingness to ration care u QALY concept and unwillingness to use it in US
Reinhardt et al (2004): Why is US spending so high, and can we afford it?
Reinhardt et al (2004): Why is US spending so high, and can we afford it?
u Ability and willingness to pay u High GDP per capita in US
u Distribution of market power and prices u Salaries of healthcare professional
u Fragmentation of purchasers
u Capacity of healthcare system u Supply side: “If you build it they will come”
u Administrative costs u Extremely complex and fragmented system
u Unwillingness to ration care u QALY concept and unwillingness to use it in US
Expenditures and Income
u As income increases, the percentage spent on basic goods and services declines—leaving more to spend on healthcare: health is a luxury good.
u This suggests that the income elasticity of demand for health care is greater than 1
An International Comparison of Per Capita Income and Health Spending
Source: OECD Statistics (October 2006)
Austr alia Austr ia
Belgi um Ca nada
Czech Re publi c
Den mar k
Finlan d
France G erma ny
Gre ece
Hun gary
Icelan d
Irel and
Italy Japan
Korea
Luxembo urg
Me xico
Nether lands
New Zealan d
Nor way
Pola nd
P ortu gal
Slo vak Re publi c
S pain
S weden
Switzerl and
Turkey United King dom
United State s
5 10
15
% G
D P
fo r
H C
0 20000 40000 60000 GDP/Capita
GDP/Capita vs. GDP HC
Gap Exceeds Expectation
Income Elasticity • Across countries, the elasticity of health spending
with respect to income is greater than one
• However, the US is still an outlier after accounting for income differences
• Rather than looking across countries, what if we look at changes in income within the US?
• How does health spending change if income in one state (or city) rises faster than income in another state?
• What econometric model might you use to answer this question?
Estimating the Income Elasticity of Health Spending u Suppose are interested in estimating the elasticity of health spending with
respect to income using variation across cities in the US
u A naïve regression model might be:
u Where Hjt is the average household medical spending in city j in year t
u αj is a vector of binary variables for each city (fixed city-effects)
u γt is a vector of binary variables for each year (fixed year-effects)
u Yjt is the average household income in city j in year t
u Xjt includes a bunch of other control variables that change over time (eg population size, age distribution, insurance coverage)
u How would you interpret β ?
u Why might β be a biased estimate in this model?
log(H jt)=α j +γt +β log(Yjt)+φXjt +ε jt
Estimating the Income Elasticity of Health Spending
u The key to avoiding bias in this model is to use a source of exogenous variation in income that is not correlated with health or health spending except through its effects on income
u Instrumental variables (IV) regression (two-stage least squares is one such model)
u Acemoglu et al (2013) use exogenous income shocks due to changes in oil prices
u The idea behind the model is that some local economies in the US are heavily dependent on oil as an input of production
u In those areas, a large increase in the global price of oil can have a negative effect on growth and worker incomes
u Use IV to estimate the income elasticity of healthcare spending in the US is 0.7 (SE 0.2)
Summary: Does Income Explain Why the US Spends So Much on Health?
u Theory suggests that the share of income devoted to health should increase as the level of income increase u That is, the elasticity of medical spending with respect to income
> 1 u Estimates based on comparisons across countries suggest this is true,
but the US is still an outlier based on this comparison u Estimates of the relationship between health spending and income
within the US suggest that as cities or regions become richer they spend a smaller share of income on health u Acemoglu et al 2013 estimate that less than 5% of the growth of
health spending in the US between 1960-2005 was due to the growth in income
u Does it make sense that these two estimates are different? Why?
Reinhardt et al (2004): Why is US spending so high, and can we afford it?
u Ability and willingness to pay u High GDP per capita in US
u Distribution of market power and prices u Salaries of healthcare professional
u Fragmentation of purchasers
u Capacity of healthcare system u Supply side: “If you build it they will come”
u Administrative costs u Extremely complex and fragmented system
u Unwillingness to ration care u QALY concept and unwillingness to use it in US
Physician Earnings
Supplier Concentration
u Suppliers (hospitals, physicians, pharmaceutical manufacturers) can often have substantial market power in negotiating with purchasers (insurance companies, consumers, the government)
u Question: To what extent is market power among suppliers increasing prices?
u Capps and Dranove (2004) aggregate results from several studies, and estimate that a 15% increase in hospital Herfindahl Herschman Index (HHI) in a metro market tends to lead to a 10% increase in prices
u However, prices in the US are still much higher even in fairly competitive markets with many hospitals and physicians
Reinhardt et al (2004): Why is US spending so high, and can we afford it?
u Ability and willingness to pay u High GDP per capita in US
u Distribution of market power and prices u Salaries of healthcare professional
u Fragmentation of purchasers
u Capacity of healthcare system u Supply side: “If you build it they will come”
u Administrative costs u Extremely complex and fragmented system
u Unwillingness to ration care u QALY concept and unwillingness to use it in US
66
Reinhardt et al (2004): Why is US spending so high, and can we afford it?
u Ability and willingness to pay u High GDP per capita in US
u Distribution of market power and prices u Salaries of healthcare professional
u Fragmentation of purchasers
u Capacity of healthcare system u Supply side: “If you build it they will come”
u Administrative costs u Extremely complex and fragmented system
u Unwillingness to ration care u QALY concept and unwillingness to use it in US
Administrative Costs
u The medical system in the US is extremely fragmented and complicated
u Whereas in some countries there is one public agency that determines prices and deals with all payments, in the US we have thousands of independent private companies doing this u Each hospital, physician, insurer, and pharmacy pair
has to negotiate their own price, and come up with their own system of processing payments
u Could be inefficient if there are economies of scale for either insurers or providers
Administrative Costs
u Can be very hard to measure u Sensitive to assumptions about what is actually
administrative
u Reinhardt et al cite a study that estimated administrative costs to be at least 24% of total US health spending in 1999
u Based on the NHE accounting discussed last class, the “official” percentage in 2005 was 7.2%, which is probably too low
u Based on what we have learned so far, do you think administrative costs can explain why the US is so different?
Administrative Complexity?
Source: Kaiser Family Foundation
Reinhardt et al (2004): Why is US spending so high, and can we afford it?
u Ability and willingness to pay u High GDP per capita in US
u Distribution of market power and prices u Salaries of healthcare professional
u Fragmentation of purchasers
u Capacity of healthcare system u Supply side: “If you build it they will come”
u Administrative costs u Extremely complex and fragmented system
u Unwillingness to ration care u QALY concept and unwillingness to use it in US
Can We Afford It?
Depressing Wage Growth?
Depressing Wage Growth?
Source: Bosworth and Perry (1994)
What Factors Drive US Healthcare Spending u Ability and willingness to pay
u High GDP per capita in US
u Distribution of market power and prices u Salaries of healthcare professional u Fragmentation of purchasers
u Capacity of healthcare system u Supply side: “If you build it they will come”
u Administrative costs u Extremely complex and fragmented system
u Unwillingness to ration care u QALY concept and unwillingness in US to base
medical decisions on it (eg Medicare)
