Competition in healthcare markets
Competition in healthcare is fiercely debated…
- Competition has become the pariah of healthcare concepts. Only a few still want to use the
word
- The coronavirus shows that we cannot leave healthcare to the market
- In healthcare, competition is an easy scapegoat. So you hear many politicians shouting that
we should get rid of that. Arguing that would solve all problems is simplistic whishful
thinking. It sounds good and scores well electorally, but it does not do justice to the complex
reality of healthcare. In the health care sector, we need a combination of market incentives
and government regulation.
o Central to industrial organization
2 parts of this course
1
,Lecture 1- Industrial organization of healthcare
Industrial organization
Industrial organization is the combination of
- market competition
- market incentive
- government regulation
‘’ To study industrial organization is to study the functioning of markets, a central concept in
microeconomics (Jean Tirolte)’’
If not working well, the public goals of healthcare cannot be satisfied
o The goals: Quality, access (coverage), affordability
IO in healthcare markets
‘’No other market of substantial importance violates the requirements of perfect competition so
radically’’ (David Dranove, Mark Satterthwaite, 2000)
Requirements of perfect competition
- Free entry and exit
o Hc: Luckily not, entry barriers
- Standardized (homogenous) products
o Hc: A lot of heterogenous products
- Price taking behavior
o Complete absence of market power
o Individual buyers/ sellers cannot influence market price
o Hc: dominant positions on both sides
- Perfect information (transparency)
o Hc: information asymmetry
Government regulation is required, but
Just like markets, governments are also far from perfect
- Hence, improving the functioning of healthcare markets is all about: how to navigate
between market failure & government failure (=IO)
Industrial organization of healthcare markets: key issues (Martin
Gaynor, Kate Ho, Robert Town, 2015)
- “There has been growing interest among economists in recent years in the industrial
organization of health care markets. This is due in part to the growing prominence of health-
care markets in policy issues, the increasing availability of rich datasets on health care,
advances in economics methodology, and institutional changes that have led to a greater
role and prominence for markets in health care.”
Multistage model (Gaynor et al. 2015)
Discussion of literature structured around 5 stages:
1. Quality determination in provider markets M1
2. Price and network determination in provider markets M2
3. Premium determination in insurance markets M4
4. Consumer choice in insurance markets M4
5. Incentives and provider referral decision’s/consumer utilization M3
2
, Each stage has impact on equilibrium outcome & welfare
Stages are related: “optimal choices in one stage are functions of expectations regarding the
rest”
Industrial organization of healthcare: quality determination in
provider markets
Quality measures
- Inpatient mortality: indicator for measuring quality differences between hospitals
o How many receive treatment and how many die
- Hip replacements
o Readmission
- PROMS: patient reported outcome measures
Measure must be related to hospital activities
o Length of stay: no quality measure
Hospital competition & quality: theory
- What mechanism could explain that hospital competition affects quality?
o Strategic choice for lower quality is unlikely
o More likely: lower effort in more concentrated hospital markets
Markets with regulated prices
- zej = eq. quality hospital j
- p = regulated price
- cq & cz = mc of quantity & quality
- sj = market share hospital j
- D = total market demand
- Quality is increasing in price, the elasticity of demand with respect to quality, and the firm’s
total demand.
o Higher price> higher quality
o More elastic demand to quality> higher quality in competitive markets> attract more
patients
o If elasticity is low: less incentive to increase quality
- Quality is decreasing in the marginal costs of quantity or quality
o If regulated price is high enough > people respond to quality differences > hospitals
have incentive to compete on quality > more competition will result in more quality
Markets where hospitals set prices & quality (NL)
- z = quality
- p = price
- d = mc of quality
- z = quality elasticity of demand
- p = price elasticity of demand
- Quality will increase if the quality elasticity of demand increases or the price elasticity of
demand declines (and vice versa)
o It is ambiguous what will happen to quality if competition increases
Depends on what is most important to the buyers of healthcare
3
, If patients more responsive to price than to quality> will lead to lower
prices, lower quality
If quality elasticity of demand is higher than price elasticity> quality
improvements
- Quality will also increase if price increases relative to the marginal cost of quality (and fall if
the opposite happens)
Hospital competition & quality: empirics
SCP (Structure-conduct-performance)
- Linear regression:
o Look at performance> model conduct> you use an indicator HHI
- HHI max: 10.000 (100^2)
o Sum of squared market shares
o Closer to 10.000, the more concentrated the market, less competitive
Event study/ DiD
- Superior to SCP, but requires more data
- Equation usually estimated
o p = price
o q = quantity
o XD = demand shifters
o W = cost shifters
o HHI = market concentration
- Methodological problems: Each study has its advantages and disadvantages
Markets with regulated prices
- Limitation: these studies use AMI data. Problem: AMI is very specific, how do conclusions
hold in other markets
- For example, Kessler & McClellan (2000, QJE)
o SCP framework
o Significantly higher risk-adjusted 1-year mortality for Medicare AMI patients in more
concentrated markets
1. Risk adjusted:
1. Important because the risk (healthy/ sick)/ case-mix of patients
should be considered.
2. Otherwise incentive for adverse selection
- For example, Gaynor et al. (2013, AEJ:EP) NHS after 2006 reform (fixed prices after this)
o Combination of SCP with ‘natural experiment’ DiD-approach
o Risk-adjusted mortality from AMI fell more at hospitals in less concentrated markets
than at hospitals in more concentrated markets
Limitations: almost all studies use data of AMI markets
Markets where hospitals set prices & quality
- Limited number of studies
Propper et al. (2008)> NHS internal market 1991-1999
- DiD
- Differences in mortality for hospitals in areas with competitors (T) vs. those with no
competitors were higher during period when competition was promoted (1991-1995) (T-1)
compared to when competition was discouraged (1996-1998)(T+1)
4
Competition in healthcare is fiercely debated…
- Competition has become the pariah of healthcare concepts. Only a few still want to use the
word
- The coronavirus shows that we cannot leave healthcare to the market
- In healthcare, competition is an easy scapegoat. So you hear many politicians shouting that
we should get rid of that. Arguing that would solve all problems is simplistic whishful
thinking. It sounds good and scores well electorally, but it does not do justice to the complex
reality of healthcare. In the health care sector, we need a combination of market incentives
and government regulation.
o Central to industrial organization
2 parts of this course
1
,Lecture 1- Industrial organization of healthcare
Industrial organization
Industrial organization is the combination of
- market competition
- market incentive
- government regulation
‘’ To study industrial organization is to study the functioning of markets, a central concept in
microeconomics (Jean Tirolte)’’
If not working well, the public goals of healthcare cannot be satisfied
o The goals: Quality, access (coverage), affordability
IO in healthcare markets
‘’No other market of substantial importance violates the requirements of perfect competition so
radically’’ (David Dranove, Mark Satterthwaite, 2000)
Requirements of perfect competition
- Free entry and exit
o Hc: Luckily not, entry barriers
- Standardized (homogenous) products
o Hc: A lot of heterogenous products
- Price taking behavior
o Complete absence of market power
o Individual buyers/ sellers cannot influence market price
o Hc: dominant positions on both sides
- Perfect information (transparency)
o Hc: information asymmetry
Government regulation is required, but
Just like markets, governments are also far from perfect
- Hence, improving the functioning of healthcare markets is all about: how to navigate
between market failure & government failure (=IO)
Industrial organization of healthcare markets: key issues (Martin
Gaynor, Kate Ho, Robert Town, 2015)
- “There has been growing interest among economists in recent years in the industrial
organization of health care markets. This is due in part to the growing prominence of health-
care markets in policy issues, the increasing availability of rich datasets on health care,
advances in economics methodology, and institutional changes that have led to a greater
role and prominence for markets in health care.”
Multistage model (Gaynor et al. 2015)
Discussion of literature structured around 5 stages:
1. Quality determination in provider markets M1
2. Price and network determination in provider markets M2
3. Premium determination in insurance markets M4
4. Consumer choice in insurance markets M4
5. Incentives and provider referral decision’s/consumer utilization M3
2
, Each stage has impact on equilibrium outcome & welfare
Stages are related: “optimal choices in one stage are functions of expectations regarding the
rest”
Industrial organization of healthcare: quality determination in
provider markets
Quality measures
- Inpatient mortality: indicator for measuring quality differences between hospitals
o How many receive treatment and how many die
- Hip replacements
o Readmission
- PROMS: patient reported outcome measures
Measure must be related to hospital activities
o Length of stay: no quality measure
Hospital competition & quality: theory
- What mechanism could explain that hospital competition affects quality?
o Strategic choice for lower quality is unlikely
o More likely: lower effort in more concentrated hospital markets
Markets with regulated prices
- zej = eq. quality hospital j
- p = regulated price
- cq & cz = mc of quantity & quality
- sj = market share hospital j
- D = total market demand
- Quality is increasing in price, the elasticity of demand with respect to quality, and the firm’s
total demand.
o Higher price> higher quality
o More elastic demand to quality> higher quality in competitive markets> attract more
patients
o If elasticity is low: less incentive to increase quality
- Quality is decreasing in the marginal costs of quantity or quality
o If regulated price is high enough > people respond to quality differences > hospitals
have incentive to compete on quality > more competition will result in more quality
Markets where hospitals set prices & quality (NL)
- z = quality
- p = price
- d = mc of quality
- z = quality elasticity of demand
- p = price elasticity of demand
- Quality will increase if the quality elasticity of demand increases or the price elasticity of
demand declines (and vice versa)
o It is ambiguous what will happen to quality if competition increases
Depends on what is most important to the buyers of healthcare
3
, If patients more responsive to price than to quality> will lead to lower
prices, lower quality
If quality elasticity of demand is higher than price elasticity> quality
improvements
- Quality will also increase if price increases relative to the marginal cost of quality (and fall if
the opposite happens)
Hospital competition & quality: empirics
SCP (Structure-conduct-performance)
- Linear regression:
o Look at performance> model conduct> you use an indicator HHI
- HHI max: 10.000 (100^2)
o Sum of squared market shares
o Closer to 10.000, the more concentrated the market, less competitive
Event study/ DiD
- Superior to SCP, but requires more data
- Equation usually estimated
o p = price
o q = quantity
o XD = demand shifters
o W = cost shifters
o HHI = market concentration
- Methodological problems: Each study has its advantages and disadvantages
Markets with regulated prices
- Limitation: these studies use AMI data. Problem: AMI is very specific, how do conclusions
hold in other markets
- For example, Kessler & McClellan (2000, QJE)
o SCP framework
o Significantly higher risk-adjusted 1-year mortality for Medicare AMI patients in more
concentrated markets
1. Risk adjusted:
1. Important because the risk (healthy/ sick)/ case-mix of patients
should be considered.
2. Otherwise incentive for adverse selection
- For example, Gaynor et al. (2013, AEJ:EP) NHS after 2006 reform (fixed prices after this)
o Combination of SCP with ‘natural experiment’ DiD-approach
o Risk-adjusted mortality from AMI fell more at hospitals in less concentrated markets
than at hospitals in more concentrated markets
Limitations: almost all studies use data of AMI markets
Markets where hospitals set prices & quality
- Limited number of studies
Propper et al. (2008)> NHS internal market 1991-1999
- DiD
- Differences in mortality for hospitals in areas with competitors (T) vs. those with no
competitors were higher during period when competition was promoted (1991-1995) (T-1)
compared to when competition was discouraged (1996-1998)(T+1)
4
