Regul egulat ator ory y gaps ps wi withi thin n healthc he althcar are e fi fina nancing ncing in in th the e Sou outh th Af Afri rican can pri rivat ate e se sect ctor or Health alth Mark rket et Inquir quiry y Semina nar 1 Fe 1 Februar uary 2018 18
We are not where we were meant to be Medical schemes operate in an unbalanced incomplete regulatory framework. We have stopped part way through point 3 on the intended trajectory. Purpose of solidarity based reforms was to prevent the industry excluding those in need of care. Source: Ministerial Task Team on SHI, July 2005 Taken from IPASA research at https://goo.gl/xkR8DY 2
“ South Africa is unusual in having open enrolment and community rating without risk equalisation. This was not a policy oversight , but a question of timing, and the South African Department of Health considers that the environment is now ready for the introduction of a Risk Equalisation Fund (REF).” Source: Prof Heather McLeod, 2005. (Emphasis added) Quoted by Minister of Health in the same year Given the time that has passed, this point can be recharacterised as indeed being a policy oversight.
That the current regulatory environment permits anti selection should not be in question, but the extent and effect of the behavior is subject to some debate. 4
Three levels of anti-selection 1 Anti-selection into and out of the medical scheme environment 2 Anti-selection between schemes 3 Anti-selection between benefit options 5
Anti-selection into and out of the 1 medical scheme environment 12% 12,0% 2002 2016 10,0% 10% Medical scheme membership profile 8,0% 8% 6,0% 4,0% 6% 2,0% 4% 0,0% < 1 1-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+ 2% Total population 0% Current medical scheme <01 01-04 05-09 10-14 75+ 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 Population earning above the tax threshold The dip and the heavy tail in scheme membership Its getting worse over time which increases indicate selective behavior. Both increase the cost base costs over time. We estimate 1.3% - 1.9% pa 6 Stats SA, CMS data
Anti-selection into and out of the 1 medical scheme environment 70% 30% 60% 25% 50% n covered 20% Perentage female 40% tion 15% roportio 30% 10% Pro 20% Total population 5% 10% Current medical scheme 2002 2016 Population earning above the tax threshold 0% 0% 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75+ 5-9 10-14 15-19 70-74 75-79 Under 1 1-4 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 80-84 85+ While coverage overall is around 16% of the Behaviour goes beyond ‘age only’. Maternity population, its closer to 25% for those over 40. selection is good needs-based example 7 Stats SA, CMS data
Anti-selection into and out of the 1 medical scheme environment 90% 80% 70% Proportion covered ‘Dip’ in membership seen even in higher 60% 50% deciles, which suggests selection, 40% compounded by affordability. 30% 20% 10% 0% 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75+ Decile 7 Decile 8 Decile 9 Decile 10 8 Stats SA
Anti-selection into and out of the 1 medical scheme environment Average Average Average Long term trend of open schemes needing Annual GCI Annual GCI Annual GCI higher contribution increase than restricted Increase pbpa Increase pbpa Increase pbpa (2000 - 2016) (2000 - 2006) (2007 - 2016) schemes, including pre-GEMS, whilst in the All open schemes 9.8% 11.4% 8.9% All restricted schemes 8.0% 8.3% 7.9% same operating environment. Noting that Open schemes excluding 9.8% 10.8% 9.2% open schemes also have some employer DHMS Restricted schemes excluding 8.0% 8.3% 7.8% based membership. GEMS In 2017 restricted schemes had roughly 15% lower contributions and 5% richer benefits. 9 CMS data
Anti-selection into and out of the 1 medical scheme environment Mitigation options: Some level of compulsion or means based Positive solidarity fine for non participation. Eases burden on public sector Mandated income cross subsidy Directionally towards NHI Employer subsidies (noting family size issues) Penalises those who need care More stringent penalties for late joiners Penalties at the time seem not to be Stronger underwriting effective 10
2 CMS data Average age versus industry average (15,00) (10,00) 20,00 25,00 10,00 15,00 (5,00) 5,00 - Anti-selection between schemes Spectramed Selfmed Keyhealth Resolution Fedhealth Cape Medical Plan Medshield Topmed Compcare Scheme risk profiles vary widely, which translates to different levels of claims. Medihelp Open Bestmed Open schemes worse on average. Perilous to ignore income dynamics. Suremed Health Sizwe DHMS Bonitas Genesis Momentum Hosmed Medimed Makoti Thebemed Transmed Parmed De Beers BP Sedmed WITS Anglo Tiger Brands Profmed Rhodes UKZN Anglovaal AECI Engen Grintek Lonmin Medipos Metropolitan SABC Motohealh Golden Arrow Quantum Nedgroup Barloworld Malcor PG Group Old Mutual <table of Mcleod> Restricted Platinum Health Alliance-Midmed Libcare Rand Water TFG SAMWUMed Sisonke Bankmed Sasolmed GEMS Wooltru Pick n Pay CAMAF Massmart Witbank Netcare Impala SABC Naspers LA-Health Retail Imperial Remedi MBMed Umvuzo BMW Horizon POLMED Tsogo Sun Fishmed Glencore 11
Anti-selection between schemes 2 250% Income cross subsidies within, and between schemes are important. 200% Generally, restricted schemes are Loss Ratio 150% better able to cross subsidise on income. Previous reform pathways 100% included consideration of broader 50% income cross subsidies. 0% 0 5 000 10 000 15 000 20 000 25 000 Grouped income band 24 schemes, 40 options, 3m lives, overall loss ratio 93% 12
Anti-selection between schemes 2 Mitigation options: Risk Equalisation or virtual pooling, along Balance quick, easy and practical with what’s with income cross subsidies necessary Stronger underwriting when moving Taking care to avoid unintended harm to between schemes those in need of care Sequencing of any such reforms are important so as to avoid adverse consequences 13
Anti-selection within schemes 3 Illustrative example of option selection dynamics that are not evident when looking at high level data only. 14 CMS data
Anti-selection within schemes 3 40% Percentage of medical scheme beneficaireis 35% 30% Buydown behavior is evident over 25% time, and costs schemes 1-2% per 20% annum. Effect is higher for open 15% schemes than restricted schemes. 10% 5% 0% 2008 2009 2010 2011 2012 2013 2014 2015 1 2 3 4 Benefit option quartiles 15 CMS data
Anti-selection within schemes 3 Downgrades 1 600 1 400 Drop in claims post 1 200 Claims PLPM, inflation-adjusted downgrade less than the 1 000 drop in contribution 800 600 Time of change 400 200 0 -36 -33 -30 -27 -24 -21 -18 -15 -12 -9 -6 -3 0 3 6 9 12 15 18 21 24 27 30 33 36 Months 16 Sample of Insight data, 2016
Anti-selection within schemes 3 Upgrades 2 500 2 000 Sharp increases in claims (for Claims PLPM, inflation-adjusted PMB and non PMB claims) 1 500 post upgrade, suggests 1 000 selective behavior. Time of change 500 0 -36 -33 -30 -27 -24 -21 -18 -15 -12 -9 -6 -3 0 3 6 9 12 15 18 21 24 27 30 33 36 Months 17 Sample of Insight data, 2016
Anti-selection within schemes 3 130,0% 120,0% 110,0% Loss ratio impact in first year of option 100,0% 90,0% change. Both upgrades and 80,0% downgrades leave a scheme worse off. 70,0% 60,0% 50,0% Upgrade Downgrade Before After 18 Sample of Insight data, 2016
Comments on other risk pooling issues raised Variety, choice and innovation trade off against solidarity and cross subsidy. Restricted schemes offer less choice, greater solidarity and better cross subsidy; Open schemes offer more choice and innovation. Important to get the balance right based on societal objectives. Self sustaining options are a self inflicted fragmentation of risk pools and unnecessary. The requirement could be relaxed or removed, with the CMS continuing to monitor scheme subsidies. Circular 8 of 2006 may be viable subject to allowing for income subsidies. Benefit designs should be more directly comparable through some standardized templates. 19
Recommend
More recommend
