r r Ps rt q - - PowerPoint PPT Presentation
r r Ps rt q r P sst
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
Figure 3: Federal Comprehensive Risk-Based Medicaid Managed Care Expenditures, Total and as a Percentage of Overall Federal Medicaid Expenditures, Fiscal Years 2004-2014
Note: For this analysis, expenditures were adjusted for inflation using the gross domestic product price index to 2014 dollars.
Federal expenditures for managed care varied widely by state—ranging
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
5.40 16.00 43.39 38.00 28.32 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 Marketplaces Medicare Medicaid $22 $156 $162 $362 $314 $0 $100 $200 $300 $400 $500 $600 Marketplaces Medicare Medicaid
Enrollment (millions) Expenditure (billions)
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✻ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✻ ✴ ✹✸
■♥tr♦❞✉❝t✐♦♥ ▼♦t✐✈❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✻ ✴ ✹✸
❇❛❝❦❣r♦✉♥❞ ❛♥❞ ❉❛t❛ ❇❛❝❦❣r♦✉♥❞
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✼ ✴ ✹✸
❇❛❝❦❣r♦✉♥❞ ❛♥❞ ❉❛t❛ ❇❛❝❦❣r♦✉♥❞
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✽ ✴ ✹✸
❇❛❝❦❣r♦✉♥❞ ❛♥❞ ❉❛t❛ ❇❛❝❦❣r♦✉♥❞
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✽ ✴ ✹✸
❇❛❝❦❣r♦✉♥❞ ❛♥❞ ❉❛t❛ ❇❛❝❦❣r♦✉♥❞
◆❡t✇♦r❦ ❱❛r✐❛t✐♦♥
Sample Recipients Provider Network For- Provider Non- Physician Hospital Plan Assigned Profit Sponsored SSI SSI (%) (%) Affinity
1,322 0.63 0.79 Amerigroup
1,190 0.71 0.65 HIP 5,760 545 0.74 0.96 Health First
846 0.76 0.72 Health Plus
1,049 0.74 0.99 Metroplus
1,175 0.55 0.49 Fidelis
1,295 0.67 0.77 Neighborhood
1,295 0.81 0.93 United
1,125 0.74 0.92 Wellcare
149 0.55 0.67 Overall 102,862 9,963 0.70 0.78
Notes: This table lists the 10 MCO plans competing for enrollees in NYC Medicaid. Medicaid beneficiaries who
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✾ ✴ ✹✸
❇❛❝❦❣r♦✉♥❞ ❛♥❞ ❉❛t❛ ❇❛❝❦❣r♦✉♥❞
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✵ ✴ ✹✸
❇❛❝❦❣r♦✉♥❞ ❛♥❞ ❉❛t❛ ❇❛❝❦❣r♦✉♥❞
Qualify for Medicaid 90-day opt out period 9-month lock-in begins FFS Medicaid MCO Medicaid 30/60-day active choice period
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✶ ✴ ✹✸
❊♠♣✐r✐❝❛❧ ❙tr❛t❡❣② ■❞❡♥t✐☞❝❛t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✷ ✴ ✹✸
❊♠♣✐r✐❝❛❧ ❙tr❛t❡❣② ■❞❡♥t✐☞❝❛t✐♦♥
✶✵
✶✵
✶✵
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✸ ✴ ✹✸
❊♠♣✐r✐❝❛❧ ❙tr❛t❡❣② ■❞❡♥t✐☞❝❛t✐♦♥
20% 40% 60% 80% 100% Fraction of Months in Plan Affinity Amerigroup Health First Health Plus HIP Metroplus Neighborhood Fidelis United Wellcare
In Plan | Assigned to Plan In Plan | Not Assigned to Plan
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✹ ✴ ✹✸
❊♠♣✐r✐❝❛❧ ❙tr❛t❡❣② ■❞❡♥t✐☞❝❛t✐♦♥
F-Test P-value Baseline Auto Self Mean Assigned Selected (1) (2) (3) Age 37.462 [0.656] [0.000]∗∗ (13.807) Male 0.399 [0.054]† [0.000]∗∗ (0.490) Black 0.285 [0.560] [0.000]∗∗ (0.451) Baseline Office Spending 33.891 [0.375] [0.000]∗∗ (150.862) Baseline OPD Spending 38.674 [0.467] [0.000]∗∗ (143.230) Baseline Clinic Spending 24.162 [0.457] [0.000]∗∗ (115.524) Baseline Inpatient Spending 156.306 [0.161] [0.000]∗∗ (1152.549) Baseline Office Quantity 0.281 [0.941] [0.000]∗∗ (0.642) Baseline OPD Quantity 0.218 [0.415] [0.000]∗∗ (0.743) Baseline Clinic Quantity 0.199 [0.497] [0.000]∗∗ (1.025) Baseline Inpatient Spending 0.019 [0.050]† [0.000]∗∗ (0.093) Observations 102,862 102,862
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✺ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✻ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.2 .4 W e l l c a r e A m e r i g r
p U n i t e d H I P A f f i n i t y N e i g h b
h
F i d e l i s H e a l t h P l u s H e a l t h F i r s t M e t r
l u s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✼ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
(1) (2) Red Form Inst Var Amerigroup
(0.032) (0.038) United
(0.033) (0.038) Wellcare
(0.073) (0.092) HIP
(0.034) (0.039) Affinity
(0.034) (0.040) Neighborhood
(0.032) (0.038) Health First 0.139∗∗ 0.156∗∗ (0.046) (0.051) Health Plus 0.084∗ 0.095∗ (0.034) (0.040) Metroplus 0.087∗∗ 0.102∗∗ (0.031) (0.036) Observations 900443 900443
educed form and IV results of health plan assignment on
❈♦♥t❡①t
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✽ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✾ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✶✾ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.5 20 40 60 80 100 quantile Coefficient 95% Confidence Interval ❢♦❧❧♦✇✐♥❣ ❈❤❡t✈❡r✐❦♦✈✱ ▲❛rs❡♥ ❛♥❞ P❛❧♠❡r ✭✷✵✶✻✮
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✵ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.2 .4 .6 .8 20 40 60 80 100 quantile Coefficient 95% Confidence Interval ❢♦❧❧♦✇✐♥❣ ❈❤❡t✈❡r✐❦♦✈✱ ▲❛rs❡♥ ❛♥❞ P❛❧♠❡r ✭✷✵✶✻✮
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✶ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✷ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.05 .1 Log Spending Relative to Fidelis High-cost Low-cost Price Quantity
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✸ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.05 .1 Log Spending Relative to Fidelis High-cost Low-cost Price Quantity
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✸ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✹ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
T
a l O f f i c e
C P O f f i c e
p e c O P E D I P R x
.1 .2 High Cost Low Cost
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✺ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.46 .19 .1 .066 .064 .053 .05 .014 .013
.1 .2 .3 .4 .5 Pct of Diff in Tot Spend btw High and Low Cost Plans I n p a t i e n t R x C l i n i c O u t p a t i e n t D e n t a l L a b O f f i c e , S p e c i a l t y T r a n s p
t O f f i c e , P C P E D O t h e r
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✻ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✼ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✼ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✼ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✼ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.55 .6 .65 .7 .75 .8 Physician Network Size
.1 .2 Spending Effect Estimate .5 .6 .7 .8 .9 1 Hospital Network Size
.1 .2 Spending Effect Estimate
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✽ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.55 .6 .65 .7 .75 .8 Physician Network Size
.1 .2 Spending Effect Estimate .5 .6 .7 .8 .9 1 Hospital Network Size
.1 .2 Spending Effect Estimate
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✽ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.55 .6 .65 .7 .75 .8 Physician Network Size
.1 .2 Spending Effect Estimate .5 .6 .7 .8 .9 1 Hospital Network Size
.1 .2 Spending Effect Estimate
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✽ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.2 Over 40 Under 40 High Cost Low Cost
Old vs. Young
.1 .2 Not Black Black High Cost Low Cost
Not Black vs. Black
.1 .2 Female Male High Cost Low Cost
Female vs. Male
.1 .2 High Spending Low Spending High Cost Low Cost
High vs. Low Baseline Spending
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✾ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.2 Over 40 Under 40 High Cost Low Cost
Old vs. Young
.1 .2 Not Black Black High Cost Low Cost
Not Black vs. Black
.1 .2 Female Male High Cost Low Cost
Female vs. Male
.1 .2 High Spending Low Spending High Cost Low Cost
High vs. Low Baseline Spending
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✷✾ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✵ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✶ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✷ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✷ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.05
Willingness-To-Stay
.1 .2
Spending
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✸ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.2 .4 W e l l c a r e A m e r i g r
p U n i t e d H I P A f f i n i t y N e i g h b
h
F i d e l i s H e a l t h P l u s H e a l t h F i r s t M e t r
l u s 100 200 300 400 Total Enrollment (thousands) W e l l c a r e A m e r i g r
p U n i t e d H I P A f f i n i t y N e i g h b
h
F i d e l i s H e a l t h P l u s H e a l t h F i r s t M e t r
l u s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✹ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
.05 Willingness-to-Stay
.1 .2 Spending Effect Estimate High baseline spending High baseline spending Low baseline spending Low baseline spending
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✺ ✴ ✹✸
❘❡s✉❧ts ❙♣❡♥❞✐♥❣ ❉✐☛❡r❡♥❝❡s
♠♦r❡
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✻ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✼ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✽ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✾ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✾ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✾ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✾ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✸✾ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
.2 .4 W e l l c a r e A m e r i g r
p U n i t e d H I P A f f i n i t y N e i g h b
h
F i d e l i s H e a l t h P l u s H e a l t h F i r s t M e t r
l u s
10% 20% 30%
Fraction of Months in Plan
W e l l c a r e A m e r i g r
p U n i t e d H I P A f f i n i t y N e i g h b
h
F i d e l i s H e a l t h P l u s H e a l t h F i r s t M e t r
l u s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✵ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
▼♦❞❡❧ ❋✐❣✉r❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✶ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
▼♦❞❡❧ ❋✐❣✉r❡s
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✶ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✷ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✷ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✷ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✷ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✷ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✷ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✷ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✸ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✸ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✸ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✸ ✴ ✹✸
▼♦❞❡❧ ❈♦♠♣❡t✐t✐♦♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✸ ✴ ✹✸
❆♣♣❡♥❞✐①
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✹ ✴ ✹✸
❆♣♣❡♥❞✐①
.1
MMC Spending
.1 .2
Total Spending
.05 .1
FFS Spending
.1 .2
Total Spending
❘❡t✉r♥
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✺ ✴ ✹✸
❆♣♣❡♥❞✐①
❇❛❝❦ Fidelis Affinity
.5
Assigned Hospital Network
.5 1
Assigned Physician Network
Affinity Amerigroup Health First Health Plus HIP Metroplus Neighborhood Fidelis United Wellcare
Metroplus United
Notes: This figure plots recipients’ physician and hospital network size against their assigned physician and hospital network
December 2012 in the five counties in New York City. Recipients that qualify for Medicaid based on receipt of Supplemental Security Income (SSI) are excluded (N=900,759 patient months). To construct the binned scatter plot, I separately regress assigned and actual network size against our base controls: recipient zip, indicator variables for each half year of an episode, five-year age x gender bins, race bins, plan assignment, and county x year x month of assignment. I then take the mean residuals in each plan by zip bin, adding the mean network size (actual or assigned) back in to facilitate interpretation of the results.
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✻ ✴ ✹✸
❆♣♣❡♥❞✐①
❇❛❝❦
▼♦r❡✳✳✳
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✼ ✴ ✹✸
❆♣♣❡♥❞✐①
❇❛❝❦
1 2
Avoidable Hospitalizations
.1 .2
Spending
1 2 3
Preventive Care
.1 .2
Spending
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✽ ✴ ✹✸
❆♣♣❡♥❞✐①
❇❛❝❦
1 2
Avoidable Hospitalizations
.1 .2
Spending
1 2 3
Preventive Care
.1 .2
Spending
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✽ ✴ ✹✸
❆♣♣❡♥❞✐①
❇❛❝❦ 𝑟1 𝑟2
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✹✾ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = 𝑟 Start with this region
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✵ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = 𝑟 𝑟2 = 𝑟 If insurer 1 sets 𝑟1 < 𝑟 , then insurer 2’s best response is to set 𝑟2 = 𝑟 and take all enrollees.
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✶ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = 𝑟 𝑟2 = 𝑟 𝜌1 = 0 𝜌2 = 𝑂(𝑠 − 𝑟 ) If insurer 1 sets 𝑟1 < 𝑟 , then insurer 2’s best response is to set 𝑟2 = 𝑟 and takes all enrollees.
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✷ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = 𝑟 𝑟2 = 𝑟 But, if insurer 2 sets 𝑟2 = 𝑟 , then insurer 1 can make positive profits by setting 𝑟1 = 𝑟
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✸ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = 𝑟 𝑟2 = 𝑟 But, if insurer 2 sets 𝑟2 = 𝑟 , then insurer 1 can make positive profits by setting 𝑟1 = 𝑟 and splitting the enrollees evenly 𝜌1 = 𝑂 2 (𝑠 − 𝑟 ) 𝜌2 = 𝑂 2 (𝑠 − 𝑟 )
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✹ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = 𝑟 𝑟2 = 𝑟 But, if insurer 1 sets 𝑟1 = 𝑟 then, insurer 2 can do better by setting 𝑟2 = 𝑟 + 𝜗 and taking all choosers and half of the assignees 𝜗
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✺ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = ത 𝑟 𝑟2 = ത 𝑟 But if insurer 1 sets 𝑟1 = ത 𝑟 then, insurer 2 can do better by setting 𝑟2 = ത 𝑟 + 𝜗 and taking all choosers and half of the assignees 𝜗 𝜌1 = (𝑠 − ത 𝑟) 𝑂(1 − 𝜇) 2 𝜌2 = 𝑠 − ത 𝑟 + 𝜗 𝑂(1 + 𝜇) 2
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✻ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = 𝑟 𝑟2 = 𝑟 But then insurer 1 can do better by setting 𝑟1 = 𝑟 + 2𝜗 and taking all of the choosers 𝜗
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✼ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = ത 𝑟 𝑟2 = ത 𝑟 But then insurer 1 can do better by setting 𝑟1 = ത 𝑟 + 2𝜗 and taking all of the choosers 𝜗 𝜌1 = 𝑠 − ത 𝑟 + 2𝜗 𝑂(1 + 𝜇) 2 𝜌2 = 𝑠 − ത 𝑟 + 𝜗 𝑂(1 − 𝜇) 2
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✽ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = 𝑟 𝑟2 = 𝑟 And so on… 𝜗
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✺✾ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = ത 𝑟 𝑟2 = ത 𝑟 𝜗 Until insurer 2 is better off setting 𝑟2 = ത 𝑟 than setting 𝑟2 = 𝑟1 + 𝜗 And so on… 𝑟1 = 𝑟∗
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✻✵ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = ത 𝑟 𝑟2 = ത 𝑟 𝜗 Until insurer 2 is better off setting 𝑟2 = ത 𝑟 than setting 𝑟2 = 𝑟1 + 𝜗 And so on… Call this 𝑟∗ 𝑟1 = 𝑟∗
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✻✶ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = ത 𝑟 𝑟2 = ത 𝑟 𝜗 Until insurer 2 is better off setting 𝑟2 = ത 𝑟 than setting 𝑟2 = 𝑟1 + 𝜗 And so on… Call this 𝑟∗ (𝑠 − 𝑟∗) 𝑂(1 + 𝜇) 2 = (𝑠 − ത 𝑟) 𝑂(1 − 𝜇) 2 𝑟1 = 𝑟∗
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✻✷ ✴ ✹✸
❆♣♣❡♥❞✐①
𝑟1 𝑟2 𝑟1 = ത 𝑟 𝑟2 = ത 𝑟 𝜗 Until insurer 2 is better off setting 𝑟2 = ത 𝑟 than setting 𝑟2 = 𝑟1 + 𝜗 And so on… Call this 𝑟∗ (𝑠 − 𝑟∗) 𝑂(1 + 𝜇) 2 = (𝑠 − ത 𝑟) 𝑂(1 − 𝜇) 2 𝑟1 = 𝑟∗ Then insurer 2 drops to ത 𝑟
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✻✸ ✴ ✹✸
❆♣♣❡♥❞✐①
❇❛❝❦ 𝑟1 𝑟2 𝑟1 = ത 𝑟 𝑟2 = ത 𝑟 𝜗 Until insurer 2 is better off setting 𝑟2 = ത 𝑟 than setting 𝑟2 = 𝑟1 + 𝜗 And so on… Call this 𝑟∗ (𝑠 − 𝑟∗) 𝑂(1 + 𝜇) 2 = (𝑠 − ത 𝑟) 𝑂(1 − 𝜇) 2 𝑟1 = 𝑟∗ 𝜌1 = (𝑠 − 𝑟∗) 𝑂(1 + 𝜇) 2 𝜌2 = (𝑠 − ത 𝑟) 𝑂(1 − 𝜇) 2 Then insurer 2 drops to ത 𝑟
❙✉♣♣❧② ❙✐❞❡ ■♥❝❡♥t✐✈❡s ❏✉♥❡ ✷✵✶✼ ✻✹ ✴ ✹✸