Chapter 4

Modelling the impact of the bill

4.1 An important part of the committee's deliberations on the bill focused on the efforts to model the impact of the increased Medicare levy surcharge (MLS) thresholds, or to contest the scope and accuracy of Treasury's estimates. Treasury has estimated the budgetary impact of the bill's measures due to loss of PHI membership over the period 2008–2012 (see below). The Australian Health Insurance Association (AHIA) and the Australian Medical Association (AMA) commissioned Price Waterhouse Coopers and Access Economics respectively to examine Treasury's figures and the likely effect of the bill on both the private funds and the public hospital system. Separately, iSelect have also commissioned Access Economics to examine the effect of the bill on private health insurance (PHI) dropout, subsequent premium increases and the pressure on the public hospital system. As noted earlier, Professor Deeble has conducted his own analysis. Mr Ian McAuley and Catholic Health Australia also offered insights into the task of modelling the bill's impact, although neither undertook any econometric analysis.

4.2 This chapter presents these findings. It is important to note that the modellers make various caveats about the certainty with which these findings can be made. Not all the necessary information is publicly available and the calculations relate only to the price effect of the MLS threshold changes, not the broader motivations of people for holding private health insurance.[1] Evidence presented to the committee noted that price is not the primary reason why people take out PHI. Security, peace of mind and choice of hospital and doctor rated higher than price.[2] This would imply that private health insurance is relatively price inelastic.

The 'first round' effect: Treasury's position

4.3 Treasury's budget estimates (Table 4.1) measure the 'first round' effect of the bill's measures—the number of people who will leave the private health insurance system purely as a result of the rise in the thresholds and abstracting from any subsequent increase in premiums. It calculates that over the period 2008–2012, there will be a net saving to the public purse from increasing the threshold (excluding any increase in funding for public hospitals). This is based on:

485 000 adults[3] (644 000 people) leaving private health insurance, resulting in reduced government expenditure on the private health insurance rebate of $960 million;
an ongoing cost in foregone revenue due to exempting those in the $50 000 to $100 000 income range who are not privately insured from the levy surcharge of $660 million; and
a resulting estimated net saving of $300 million (see Table 4.1).[4]

Table 4.1: Personal income tax—increasing the Medicare threshold

Revenue ($m)

	2008–09	2009–10	2010–11	2011–12
Foregone tax revenue	-	-195.0	-235.0	-230.0
Private health insurance rebate savings	232.0	236.5	245.6	245.6

Source: Budget Paper No. 2: Budget Papers 2008–09, Commonwealth of Australia, Canberra, p. 33.

4.4 Treasury told the committee that Treasury had based its modelling of the foregone tax revenue on its personal tax model.[5] In modelling the effect of the increased threshold on rebate expenses, the Australian Tax Office provided Treasury with a confidentialised sample of data containing comprehensive information on private health insurance coverage.[6] Treasury then provided the Department of Health and Ageing and the Department of Finance and Deregulation with estimates of the 2008–09 income distributions for singles and couples with PHI. From these data were derived an estimate of the number of people with PHI in the less than 65, 65–69 and over 70 age groups.[7]

4.5 Treasury has been criticised for failing to model the effect of the bill in raising premiums (offset by any further drop in membership this causes) and therefore the cost of the rebate, and the impact on the public hospital system—the 'second round effect'. However, in evidence to the committee in June and July, Treasury explained that 'normal costing conventions' do not include costings of second-round effects. It is not a requirement of the Charter of Budget Honesty that has set guidelines for budgeting since 1998. Moreover, these second round effects are difficult to quantify. At the June hearing of Estimates, Mr Ray explained that 'the reason we do not include second round effects is that generally they are highly uncertain...we have not done that modelling because we do not feel that it is easily quantifiable'.[8] Similarly, in evidence to this inquiry, the Treasury explained:

Any effects on future premiums are deemed to be second-round effects from the policy, entail a great deal of uncertainty and would be difficult to quantify[9]...there are many other factors, such as the impact of potential marketing campaigns by funds, that might impact on the future growth.[10]

4.6 Notwithstanding the merits of these arguments, there has been criticism that Treasury's 30 per cent rebate savings estimate of $960 million is overstated given they do not measure the possible increase in premiums flowing from the fallout in PHI membership. Treasury did factor into their modelling a premium increase over the forward estimates period from factors other than the bill's influence.[11]

Other perspectives on the first-round effect

4.7 Treasury's first round effect—the number of people who will initially drop out of private health insurance—has been challenged by both Pricewaterhouse Coopers (AHIA) and Access Economics (AMA). The approach of both consultancies was to recalculate the number of people who will leave private health insurance based on Treasury's 2008–09 PHI rebate savings estimate of $232 million. The committee notes that this seems a very odd method given it makes no attempt to identify an alternative estimate.

4.8 The AHIA-commissioned Price Waterhouse Coopers report takes Treasury's savings estimate for 2008–09 and calculates the likely 'first round' fallout from PHI. The report argued that the government had significantly underestimated the effect of the increased MLS thresholds on the public health system. It claimed that the government's estimated saving of $232 million in 2008–09 is the equivalent of 908 000 people (assuming the 485 000 adults each have on average 0.87 dependants). This figure seems high, given that the people most likely to drop out of the funds are young and single. This represents 9.7 per cent of the insured population.[12]

4.9 In similar vein, the Access Economics report argued that the Treasury's savings estimates are overstated in 2008–09 and understated in subsequent years. It described the $232 million saving estimate in 2008–09 as 'highly implausible' and possible only if there was a 'sudden and large exodus of PHI members' before 1 July 2008. It added, 'we do not expect that to happen'.[13] In this context, the report emphasised that the Medicare levy surcharge is only one factor in the decision to join and remain in a private fund. Access Economics argued that other important considerations may include the perceived 'parlous state' of the public hospitals and the Lifetime Health Cover arrangements which reward early and continuous health fund membership.[14]

4.10 Access Economics noted that, based on an average rebate rate of 32 per cent, Treasury's 2008–09 saving estimate is the equivalent of $720 million in lost private health insurance contributions. For the $232 million savings estimate to be realised in 2008–09, 534 000 people claiming the average rebate rate[15] would have to drop their cover by 1 July 2008. Alternatively, for Treasury's savings estimates—and drop out figure of 485 000 adults—to be consistent, those leaving private insurance must have more expensive premiums than the average. Access Economics argued that this is not likely to be the case:

...the people who might be expected to drop their cover in the first instance would be younger high income earners who have purchased cheaper PHI products...because that is cheaper than paying the surcharge. These are ...the people whose reason for joining a fund is focussed much more on tax saving than on sharing their risk or receiving benefits.[16]

4.11 The report did not forecast the number of people who are likely to drop private health insurance cover as a result of the increased surcharge threshold. It argued that not enough is known about the price elasticity of demand for private health insurance.[17]

4.12 In its August 2008 report for iSelect, however, Access Economics did attempt to make an estimate of the likely dropout from PHI. In terms of the first round effect, it estimated that 202 000 PHI policies (359 000 people) will be dropped and of these, 51 per cent will be by those under the age of 35. This was calculated by multiplying the number of policies affected by the policy change by the proportion of singles and households that hold hospital cover to avoid the MLS. The latter figure was derived from a question in the 2004–05 National Health Survey.[18]

4.13 Notably, in its August 2008 report, Access Economics estimated a higher PHI rebate savings figure over the forward estimates than that of the Treasury: $1.2 billion compared to $960 million. The corresponding estimate of MLS tax revenue lost was $693 million (compared with Treasury's estimate of $660 million). Allowing for a first round premium increase of 2.7 per cent (see paragraph 4.20) which would increase the government's rebate liability by around $388 million, Access Economics calculates a net saving to the Commonwealth for the forward estimates period of $113 million.[19]

4.14 Professor Deeble has calculated that 488 000 PHI policies (750 000 people) will leave private health insurance as a result of the higher MLS thresholds.[20] As with most other submitters, he identified the fallout to be concentrated on younger members. He also argued that private health insurance membership is more sensitive to income than price, and younger members with lower incomes are likely to take advantage of the higher MLS thresholds and leave the private system (see chapter 5).

The 'second round' effect

4.15 The 'second round' effect refers to:

the subsequent increase in premiums to compensate for the initial loss of members from the funds;
more people dropping out of private health insurance as a consequence of higher premiums; and
the number of people newly reliant on the public hospital system.

4.16 The committee received evidence from most witnesses that the initial dropout from the funds will result in premium increases and further fallout from the funds, placing greater pressure on the public hospital system. The reason is that those most likely to drop their cover initially—the young and healthy—are those cross‑subsidising private health insurance for older people under the system of community rating.[21] However there is disagreement about the quantum of the premium increase and the pressure that will be placed on the public hospital system.

The effect on premiums

4.17 The committee received various views on the extent to which premiums are likely to increase as a result of the higher MLS thresholds. In their submission to this inquiry, AHIA argued that the bill's measures may increase premiums by as much as 10 per cent. AHIA interpreted Treasury's estimates as indicating that the fallout from the funds will be between 719 000 and 913 000 people. It assumed an average premium of $1 251 per annum[22] and an average fund policyholder aged under 65. If the higher drop out figure of 913 000 is taken, and it is assumed that those exiting made no claims in the last year, the premium increase could be as high as 10.1 per cent.[23]

4.18 Professor Deeble calculates that per person covered, the average private hospital premium (including the 30 per cent rebate) is about $930. Assuming that those dropping out in the first round (750 000 people) have average premiums, private health insurers' revenue would fall by $697 million. Once the benefit savings ($225 million) are deducted, the $427 million revenue deficit (over a remaining 8.37 million people with PHI) could be covered by a 5.1 per cent increase in premiums.[24] He adds: 'the actual result would probably be less...[and] it is hard to see that as any threat to the viability of private health insurance'.

4.19 In its May 2008 report for the AMA, Access Economics arrived at a similar figure of 5 per cent, albeit with different calculations. Access Economics equated Treasury's 2008–09 savings figure of $232 million to a $700 million loss in fund revenue. It added that the corresponding reduction in benefit payments could 'be as little as $200 million'. The resulting $500 million shortfall in revenue translates into a premium increase of 5 per cent.

4.20 In its August 2008 report for iSelect—based on its own modelling rather than Treasury's estimate—Access Economics found that the initial loss of 202 000 policies will result in, on average, a premium increase of 2.7 per cent in the first full year. This, in turn, will result in a further one per cent loss of membership or nearly 40 000 policies.[25]

4.21 The committee has not received any estimate from the government about the possible impact of the higher MLS thresholds on premiums. The government was not required to measure the second round effects. However, the Deputy Secretary of the Department of Health and Ageing, Mr David Kalisch, did advise the committee that the department had done some work into the likely increase in premiums. He suggested that premiums would increase as a consequence of the bill's measures by less than 2½per cent.[26]

4.22 There was significant variation in the estimate of any increase in premiums as a result of the MS threshold changes. It was generally accepted that there would be an increase in premiums for private health insurance regardless of the changes in the MLS thresholds ('the underlying increase') and this has not been quantified. This confuses estimates of any changes particularly as some estimates totalled the estimated underlying increase and the estimated MLS increase. Other evidence just referred to the MLS related estimate.

4.23 Those estimates that clearly related to the MLS related increase tended to be around the 2.5 per cent mark.[27] Professor Deeble estimated a maximum of 5 per cent but added that the actual result would probably be less.

Committee view

4.24 In this complex area it is not possible to make a meaningful estimate of any premium increase as a result of the increase in MLS thresholds. Health funds do not want to lose membership and will presumably try to minimise premium increases, will compete strongly with each other, and will continue to drive down costs.

Impact on CPI

4.25 The committee does not anticipate that the bill will have much impact on the consumer price index. According to the 2003–04 Household Expenditure Survey, 'hospital, residual and dental insurance' was 1.7 per cent of total household expenditure and 'hospital and medical services' (which includes doctors' fees and hospital charges as well as PHI) has a weight of 2.8 per cent in the CPI.[28] Even if the PHI 'weight' alone was as high as two per cent, then the CPI impact of the bill would be around 0.1 per cent (.02 x 5 per cent).

The effect on public hospitals

4.26 Unsurprisingly, AHIA also claims that the higher MLS thresholds will have a significant effect on the public hospital system. Using Treasury's PHI dropout figure of 485 000 adults, AHIA calculates an annual additional cost on public hospitals of $234 million. Using its own 'conservative' dropout estimate of 719 000 people, AHIA estimates an annual additional cost to public hospitals of $347 million. And with their higher dropout figure of 913 000 people[29], the Association calculates an additional annual cost to public hospitals of $442 million[30] based on an estimated average hospital benefit of $484 per person under the age of 65 paid by private health funds.[31]

4.27 AHIA's submission cites the Tasmanian Government's 2008–09 budget papers, which estimate a seven per cent increase in public hospital waiting lists.[32] AHIA suggests that the state government attributes all of this increase to the fallout from private insurance as a result of the higher MLS thresholds. This is not entirely an accurate assumption. The Budget Paper seems to indicate that the MLS change is just one of several possible factors contributing to the projected increase in public hospital waiting lists.[33]

4.28 Professor Deeble has also estimated the impact of the bill on public hospitals. He makes three assumptions:

that 'private insurance patterns of service utilization and cost are replicated exactly in the public system' (ie: the same annual admission rates per person (0.162) and relative cost index (0.79));[34]
that the average public hospital cost per admission is $4 079 (2007–08);[35] and
that 750 000 people under the age of 50 will leave PHI and use the public hospital system.

4.29 He then calculates that the increased cost on public hospitals will be $391 million per annum (750 000 x 0.162 x 0.79 x 4 079).[36] This is an extra 2.1 per cent of all inpatient expenditure. A lower annual admission rate for people under 35 reduces the annual impost on public hospitals to $311 million or 1.6 per cent of all inpatient expenditure. Professor Deeble argued that the most likely figure is about $367 million annually or 1.97 per cent of inpatient expenditure.[37] He added that the net cost to governments will hardly rise at all:

...because the Commonwealth now pays significant amounts for medical services and drugs for private patients outside the private health insurance system, and gives at least a 30 per cent rebate on premiums...[38]

4.30 The committee emphasises that calculating the effect of the bill's measures on public hospitals is not as simple as adding the hospital cost of PHI dropouts to the new cost on public hospitals. A rigorous assessment of the impact of the MLS threshold increases on the public hospital system must allow for the large number of people with PHI who concurrently use the public hospital system. Dr Robyn Lawrence, Acting Director of the Western Australian Department of Health, alerted the committee to this fact. She told the committee that the department's preliminary analysis[39]:

...indicates that in 2007-08 the threshold changes could result in an additional 12 511 public patient weighted separations. This would be mainly the result of people who would otherwise have had procedures done as private patients in private hospitals instead of having them done as public patients. If the department had the capacity to provide for all this additional demand, the estimated additional costs for the public hospital system would be of the order of $53.6 million per annum...One of the key assumptions is that the people who drop out of private health insurance are the people who will have otherwise used their insurance—that is, they are the people who resulted in these hospital separations. If this is not the case, which is possible, the impact on the public hospital system may be minimal.[40]

4.31 Access Economics' report for iSelect also acknowledged the need to take into account the use of public hospitals by those currently in PHI. It noted that 'surcharge dodgers—in addition to being younger and healthier than the average—are also more likely to exercise their rights to access public hospitals'. The effect of the bill, therefore, 'is to shift their caseload only to the extent that they are now accessing private health insurance benefits'.[41]

4.32 That noted, Access Economics did anticipate a substantial shift in procedures from privately insured patients to public patients accessing the public hospital system. It argued that, following an estimated first round fallout of 202 000 policies, there will be a shift of 82 242 'in-patient episodes of care' to the public system in the first year. The second round effect will shift a further 21 166 annual episodes, while by 2012 there will be an additional 265 000 episodes of care per year shifted to the public system.[42]

4.33 However, those who leave the private health funds will have savings from no longer paying premiums and no longer being liable the MLS. This money will increase the capacity for these people to pay for private hospital care out of their own pocket.

Committee view

4.34 The committee believes there will be some impact on the public hospital system. Private hospitals tend to specialise in elective surgery procedures rather than emergencies. The Commonwealth government has announced $3.2 billion for the National Health and Hospitals Reform Plan which includes $600 million to reduce elective surgery waiting lists (see Table 4.2). The recent federal budget also provided $1 billion of immediate funding to relieve pressure on public hospitals.[43]

Table 4.2: Proposed funding for public hospitals ($ million)

	2007–08	2008–09	2009–10	2010–11
Elective surgery waiting list reduction plan	75	155	150	220
Health and Hospital Reform—COAG—Additional funding for public hospitals	500

Source: Budget Paper No. 2, 2008–09, pp 211 and 223.

Should Treasury model the second round effect?

4.35 The committee believes whilst it would be worthwhile for Treasury to model the second round effects of the bill, the assumptions required to underpin such modelling are inadequate for a rigorous analysis to be undertaken. Treasury has itself noted that 'there is a high degree of uncertainty in the impact on potential premiums in future'.[44] Treasury also indicated that modelling the impact on public hospitals 'would be better directed to the Department of Health and Ageing'.[45] The Department of Health has indicated that while it had done some modelling on the impact of the bill on premiums, it is 'still quite speculative'.[46] The committee also recognises that these estimates may compromise the government in its forthcoming discussions with the funds on premium increases.

The 'third round' effect

4.36 Access Economics defines the 'third round' effect as those people who would have taken up private health insurance as their incomes rose into ranges subject to the MLS but now will not do so as they are no longer liable for the surcharge. It estimates that the number of members dropping out from PHI in the first and second rounds will be a constant number for each of the years 2009–2012. However, the loss of members from the third round effect will increase over the period. In other words, for each year after 2008, there will be a rising number of people who would otherwise have taken up PHI if they were liable for the surcharge.[47]

Committee view

4.37 Access Economics explains this growing rate of 'non PHI uptake' in terms of rising incomes, which would have pushed an increasing proportion of taxpayers over the current thresholds, thereby inducing some to join a fund to avoid the surcharge. The committee argues that the extent of this third round effect really reflects the failure of the previous government to increase the original threshold of $50 000 per annum (leading to a form of 'bracket creep'), rather than any fault in the proposed legislation.

Ignorance, apathy and uncertainty

4.38 Several witnesses have emphasised that the effect of this legislation on the private health funds will depend on people's knowledge of the changes and, thereafter, their personal preferences and motivations. Professor Deeble identified a combination of 'ignorance, apathy and uncertainty' as potentially limiting the immediate fallout from the funds. He told the committee:

Effects will occur over a longer period because I would not expect people to be totally aware of this—it is not the sort of thing people read every day and happily devour, they learn about a thing like this once in a while—so you could expect that a large proportion of the population, despite all of the publicity, will not even know that the change has taken place. They may know when they go to see their tax accountant and he tells them that they may not have to do this any more, but nevertheless there will be a considerable lag.[48]

4.39 Even if they are aware of the change:

...some will defer, or forget to take, the necessary action (at least until tax return time) and others will be held in private insurance by the ‘Lifetime Health Cover’ rules. If they expect their income to rise in the future, not only will liability for the surcharge come back, but the cost of private insurance will be higher. The rules allow for suspension for limited periods but most contributors would not be aware of that.[49]

4.40 Along similar lines, Professor Savage cautioned:

...it cannot be assumed that all of the people...[between the old and the new thresholds] will drop their cover. This will depend on the motivation for purchasing insurance and the value that insurance provides to them. In many markets there is considerable evidence of persistence—that is, habit—in behaviour despite changes in incentives, and this is true in health insurance markets all over the world...The Lifetime Health Cover surcharge will also provide a continuing incentive for them to maintain continuous cover. Those who enrolled after 2000 and whose premiums currently include the extra loading—the age-related loading—may also maintain their cover, to take advantage of the Lifetime Health Cover policy change, where after 10 years of continuous cover they no longer have to pay the age loading.[50]

Conclusion

4.41 Estimating the effect of the bill on private health fund membership and the public hospital system is a complex task, involving assumptions about consumers' knowledge, opinions and preferences. Nonetheless, there is broad consensus among private health insurers that the MLS threshold increases will result in an initial fallout from the funds, causing their premiums to increase. To some extent, people earning less than $100 000 per annum who would otherwise have taken out PHI will no longer do so. And throughout this process, privately insured patients will shift to the public hospital system placing added pressure on its resources, particularly for elective surgery procedures. However, this must be placed in some context. Professor Deeble noted:

...the cost of the shift—which is the main thing I was concerned about—to the public hospitals would be about $360 million a year. In a system which I think last year cost $26 billion—it will be about $27 billion this year—that is trivial.[51]

4.42 This chapter has detailed various estimates of the extent to which the funds may lose members (current and prospective), premiums may increase (immediate and medium-term), and public hospitals may be faced with higher demand. The plausibility of all these estimates can be contested, depending on the underpinning assumptions one makes. Given this, the committee stresses the importance of stakeholders' views and insights into the effect of the bill. The next chapter discusses these views in detail.

Navigation: Previous Page | Contents | Next Page