Informing The Debate About Health Savings Accounts
An Examination of Some Misunderstood Issues
Click on the text of the questions below to go directly to the corresponding answer.
- Are there now 3.2 million HSAs in use?
- Do employers offering high-deductible health insurance plans and HSAs to their employees contribute to the HSAs to help their employees cover the higher out-of-pocket costs related to the high deductibles?<
- Do high-deductible plans attached to HSAs exempt preventive benefits from the deductible?
- Are individuals enrolled in high-deductible plans attached to HSAs as satisfied with their coverage as they would be under more comprehensive plans?
- Do the latest data show whether HSAs are being used by some higher-income individuals as tax shelters?
- What do the latest data show about the health status of individuals purchasing high-deductible health insurance plans that qualify for a HSA?
- Does adverse selection actually occur in health insurance markets?
- Does the experience with consumer shopping for “LASIK” eye surgery, the costs of which have fallen over time, mean HSAs could similarly lower prices for other medical services?
- Are the premiums for high-deductible plans attached to HSAs cheaper than the premiums for high-deductible plans without HSAs?
- Do data indicating slower annual premium growth among high-deductible plans attached to HSAs, as compared to other health insurance plans, mean that HSAs are working to curb health care costs for employers and individuals?
- Do HSAs allow greater patient choice of doctors and other providers?
- Would patients have greater freedom to obtain medical services under HSAs by bypassing managed care and other health-insurance coverage restrictions?
- Would physicians and other providers receive higher reimbursements under HSAs then they do under existing health insurance plans because patients with HSAs would negotiate their own payment rates with providers?
Health Savings Accounts (HSAs), the tax-favored accounts attached to high-deductible health insurance plans that were established by the 2003 Medicare prescription drug law, are controversial. The Bush Administration has proposed new tax cuts expanding HSAs, which the Treasury projects will cost $156 billion over ten years, and the House of Representatives could consider those proposals the week of June 19, which the House Leadership has slated as “Health Week.” If these proposals are debated at that time on the House floor, the controversy over HSAs will be on display.
Underlying this controversy is the assessment by many health and tax policy experts that HSAs will be disproportionately attractive to healthier, higher-income individuals, will likely be used as tax shelters, and may lead to a weakening of employer-sponsored health insurance over time. In response, some Administration officials and other HSA proponents have argued that the actual experience with HSAs over the last two years show these concerns to be unfounded. Previous Center analyses have demonstrated, however, that these claims about the early experience with HSAs rely on misleading use of selective statistics and do not stand up under scrutiny.
This brief paper uses a question-and-answer format to examine some additional issues regarding HSAs. It is intended to help inform the HSA debate.
1. Question: Are there now 3.2 million HSAs in use?
Answer: Industry estimates indicate that there were 3.2 million high-deductible health insurance plans that qualified for a HSA as of January 2006. Such figures have been cited by HSA supporters as evidence of HSAs’ growing popularity among employers and individuals. It is unclear, however, how many of these people actually have a HSA or have contributed to one; these industry estimates do not indicate to what extent people enrolled in an HSA-eligible health insurance plan actually have established HSAs. The Government Accountability Office (GAO) reports that industry officials believe that up to half of enrollees in high-deductible plans eligible for a HSA have not opened and contributed to a HSA.
2. Question: Do employers offering high-deductible health insurance plans and HSAs to their employees contribute to the HSAs to help their employees cover the higher out-of-pocket costs related to the high deductibles?
Answer: In 2006, Mercer Human Resource Consulting, a benefits consulting firm, found that more than one-third of large employers made no contribution to their workers’ HSAs. Among firms that did make contributions, the median contribution was only $100 a year, while the median deductible was $1,200.
Similarly, an annual survey of employers conducted by the Kaiser Family Foundation and the Health Research Educational Trust (HRET) found that in 2005, some 35 percent of employers offering high-deductible plans that qualified for a HSA made no contribution to their workers’ accounts. The average (as distinguished from the median) contribution by firms that made a contribution was $553 for individual coverage and $1,185 for family coverage, significantly lower than the average high deductible of $1,901 for individual coverage and $4,070 for family coverage. As the Kaiser Family Foundation/HRET study states, “employers’ contributions to the [savings accounts] are, on average, much lower than the deductible amounts, which leaves enrollees with meaningful out-of-pocket risk.” (Note: economic theory suggests that employers who reduce their overall contributions to their employees’ health insurance will increase employees’ wages in return.)
3. Question: Do high-deductible plans attached to HSAs exempt preventive benefits from the deductible?
Answer: In response to concerns that high-deductible health insurance plans can discourage use of preventive benefits, HSA supporters often note that preventive services are exempt from the high deductible. It is true that under current law, high-deductible plans attached to HSAs are allowed to exempt preventive benefits from the deductible. But there is no requirement that such plans do so, and the Kaiser/HRET survey of employers found that only 30 percent of workers covered by HSA-qualified plans in 2005 were enrolled in plans that covered any preventive benefits before the deductible was met. The other 70 percent of workers covered by such plans were enrolled in plans that covered no preventive benefits before the deductible was satisfied.
Even plans that do cover some preventive benefits before the deduction is met do not cover such services as primary care and various prescription drugs that can avoid more expensive services like hospitalization — because federal rules do not permit HSA-qualified plans to cover such services before the deductible is met.
4. Question: Are individuals enrolled in high-deductible plans attached to HSAs as satisfied with their coverage as they would be under more comprehensive plans?
Answer: Some surveys suggest that individuals may not be as satisfied with a high-deductible plan attached to an HSA as they would be under a more comprehensive health insurance plan. A survey by McKinsey & Company found that only 44 percent of members enrolled in a consumer-directed health plan (a high deductible plan attached to either an HSA or a similar account known as a Health Reimbursement Account (HRA)) were satisfied with their care, as compared to their previous health insurance plan.
Another survey, by the Employee Benefit Research Institute and the Commonwealth Fund, found that 72 percent of individuals enrolled in a comprehensive plan were extremely or very satisfied with the quality of health care, as compared to 63 percent in a consumer-directed health plan. Some four percent were not satisfied with their comprehensive plan, while 9 percent with a consumer-directed health plan were not satisfied. (It may be noted that these surveys have a limitation, in that they do not control for the reasons why such individuals may have enrolled in high-deductible plans attached to a HSA or HRA, rather than a more comprehensive plan.)
5. Question: Do the latest data show whether HSAs are being used by some higher-income individuals as tax shelters?
Answer: Tax policy experts believe that due to their substantial tax benefits, HSAs are likely to be highly attractive to many higher-income individuals as tax shelters. Not only do the tax benefits of HSAs rise with one’s income (and tax bracket), but they also provide unprecedented tax benefits. HSAs are unique in that, unlike other savings accounts into which contributions are tax deductible, both the earnings on funds held in HSAs and withdrawals from HSAs used to pay for out-of-pocket medical costs are entirely tax-free.
Not surprisingly, a recent report from the GAO indicates that some individuals with HSAs, tend to be people who view their HSAs more as investment vehicles than as a way to help pay out-of-pocket medical costs. The GAO reported that “some account holders are primarily using HSAs as a tax-advantaged savings vehicle” and that such individuals “tend to be highly compensated individuals, [and] pay for care from other, out-of-pocket sources, rather than withdraw funds from their HSA....”
6. Question: What do the latest data show about the health status of individuals purchasing high-deductible health insurance plans that qualify for a HSA?
Answer: The most recent data come from eHealthInsurance, an online health insurance broker that serves the individual market. (As noted in previous Center analyses, data on HSAs derived solely from the individual health insurance market must be used with caution. Such data are not likely to be representative of individuals with employer-based insurance who have high-deductible plans that qualify for HSAs, and these data thus do not allow one to draw conclusions about HSA enrollment overall.)
The eHealthInsurance data drawn from the individual market indicate that enrollment in high-deductible plans that qualify for HSAs has been growing fastest among younger people. The percentage of high-deductible plan purchasers who are aged 29 or younger increased from 21 percent of purchasers in 2004 to 30 percent of purchases in 2005, a large increase in a single year. This resulted in a decline in the average age of purchasers from 40 years or age in 2004 to 38 years of age in 2005. People who are younger tend to be healthier, on average.
In addition, the average deductibles that these high-deductible plans carry have increased. In 2005, the plans of more than 60 percent of people purchasing such plans carried deductibles of $3,000 or more. In 2004, less than 50 percent of the plans purchased carried deductibles that high. Because younger people tend to be healthier, on average, than older individuals, and healthier individuals tend to be more willing to choose a higher deductible than individuals in relatively poorer health, these data may indicate that the individuals purchasing HSA-eligible plans in the individual market are becoming younger and healthier on average.
7. Question: Does adverse selection actually occur in health insurance markets?
Answer: Adverse selection occurs when healthy people and less-healthy people separate into different health insurance arrangements, and the cost of insurance for the less healthy consequently rises and places such individuals at greater risk of becoming uninsured or underinsured. Numerous health policy experts and economists have expressed concern that high-deductible plans attached to HSAs pose a significant risk of adverse selection, because such plans are likely to be disproportionately attractive to healthier individuals who do not need much in the way of health care and who consequently are less concerned about the higher out-of-pocket costs required under a high-deductible plan. If healthier individuals move to high-deductible plans attached to HSAs in large numbers over time while less healthy individuals remain in lower-deductible, comprehensive plans, then significant adverse selection will result and drive up health insurance premiums for the comprehensive plans.
The danger of adverse selection is not merely a theoretical risk. For example, within the Federal Employees Health Benefits Plan (FEHBP) during the 1990s, premiums for the more generous Blue Cross/Blue Shield (BC/BS) “High Option” plan increased much faster than premiums for the less generous BC/BS “Low Option” plan because sicker federal workers disproportionately enrolled in the High Option while healthier federal workers chose the Low Option plan. This occurred despite an actuarial difference in benefits between the two plans of only $80 in 1994. Eventually, the number of federal workers enrolling in the High Option declined substantially due to the more costly premiums; the premium for the High Option plan ended up being $2,800 higher than the Low Option premium. Similarly, Harvard University experienced adverse selection when it first offered less generous HMO coverage alongside a more generous PPO plan to its workers. Healthier people disenrolled from the PPO in large numbers and chose the less generous HMO to take advantage of the lower premiums it carried. Premiums for the PPO increased so much, relative to the premiums for the HMO, that within three years, Harvard no longer offered the PPO option because the premiums for it were viewed as prohibitive.
8. Question: Does the experience with consumer shopping for “LASIK” eye surgery, the costs of which have fallen over time, mean HSAs could similarly lower prices for other medical services?
Answer: Some HSA supporters have argued that the “self-pay” market for LASIK eye surgery offers an example of how HSAs can reduce the cost of health care services. Since LASIK surgery is not typically covered by health insurance, patients have a financial incentive to shop around and get the most competitive price. That, in turn, may have helped result in lower LASIK costs over time. HSA supporters argue that individuals with high-deductible plans attached to HSAs who have not yet met their high deductible would have a similar incentive to shop around for medical services.
The Center for Studying Health System Change (CSHSC) has examined the LASIK experience to evaluate this claim. It found that the LASIK experience is not applicable here, because LASIK differs markedly from other health care services.
- LASIK is elective and non-urgent and is a discrete service that may not vary significantly in quality across providers. In addition, patients can obtain useful price information about LASIK over the phone.
- In comparison, many health care needs and their treatments are urgent and sometimes involve emergencies. Moreover, illnesses often are difficult to diagnose and may require a bundle of patient-specific treatments for which it is not easy to compare prices.
CSHSC also determined that consumer shopping showed little potential to reduce costs even among various other elective, self-pay procedures, such as in-vitro fertilization services.
9. Question: Are the premiums for high-deductible plans attached to HSAs cheaper than the premiums for high-deductible plans without HSAs?
Answer: By definition, high-deductible plans attached to HSAs have lower premiums than do the comprehensive, lower-deductible plans typically offered by employers. But high-deductible plans attached to HSAs do not appear to be any more affordable than similar high-deductible plans without HSAs, which have been available for some time in the marketplace, particularly among smaller employers and in the individual market. For example, Mercer Human Resource Consulting found that regular PPO plans with $1,000 deductibles had premium costs similar to the premiums charged for high deductible plans attached to HSAs or to HRA plans that carry similar deductibles.
10. Question: Do data indicating slower annual premium growth among high-deductible plans attached to HSAs, as compared to other health insurance plans, mean that HSAs are working to curb health care costs for employers and individuals?
Answer: Some HSA supporters have cited as evidence of the success of HSAs in controlling rising health care costs a study conducted by Deloitte Consulting showing that the annual premiums for consumer-driven health plans — high-deductible plans attached to HSAs or HRAs — grew on average by 2.6 percent in 2006, compared to 7.1 percent for all health insurance plans. A conclusion that HSAs are containing health care costs cannot, however, be drawn from these early data. As an analysis for the National Health Policy Forum notes, “one or two years of experience are not sufficient to determine whether savings are sustainable; a plan’s claims experience tends to deteriorate over time.”
Furthermore, a number of other factors may explain the lower annual growth in premiums for consumer-driven health plans, relative to other health insurance arrangements. For example, industry data indicate that the average deductibles and/or out-of-pocket maximums of high-deductible plans that qualify for HSAs have increased significantly, which would reduce the actuarial value of the health insurance plans and thereby lower the annual growth in premium costs. As noted above, among HSA-eligible plans purchased in the individual market through eHealthInsurance, the proportion of plans carrying deductibles of more than $3,000 has increased by more than 10 percentage points, to more than 60 percent of the HSA-eligible plans purchased in 2005. Other insurance industry data show that the average deductibles and/or out-of-pocket limits for HSA-qualified high-deductible plans may have increased again between 2005 and 2006 in the individual, small group, and large group markets.
In addition, to boost enrollment in a new market, insurers may be offering discounted premium rates to first-time customers who purchase high-deductible plans attached to HSAs. These lower premiums may not endure over time. This is one of the reasons the National Health Policy Forum warned that an initial year or two of experience is insufficient to draw conclusions about savings.
Finally, some evidence suggests that HSA enrollment in the individual market is growing disproportionately among younger people, who are charged lower premiums, apparently reflecting their better health. As noted above, eHealthInsurance data on the individual health insurance market for 2005 show that the percentage of high-deductible plan purchasers aged 29 or younger increased from 21 percent of purchasers in 2004 to 30 percent of purchasers in 2005, while the percentage of older purchasers fell accordingly. The eHealthInsurance data also show that the premiums that younger enrollees are charged for HSA-eligible plans purchased in the individual market are much lower than the premiums that older enrollees pay. If the population purchasing HSA-eligible plans is becoming younger and healthier, slower growth in the overall average premium would be expected. That would not necessarily indicate that HSAs are succeeding in containing health-care cost growth. Rather, it would suggest that favorable selection may be occurring, with younger and healthier individuals disproportionately moving to high-deductible plans and HSAs.
Indeed, to the extent that premium growth is slower for HSA-eligible plans, that may reflect a general anticipation by insurers of favorable risk selection — i.e., an anticipation that the people who purchase these plans will be healthier on average than had earlier been expected, perhaps based on claims experience that indicates this is occurring. 
11. Question: Do HSAs allow greater patient choice of doctors and other providers?
Answer: Some proponents of HSAs have argued that HSAs let patients have greater choice of physicians, hospitals, and other providers. Individuals enrolled in a high-deductible plan attached to a HSA can use tax-favored savings held in a HSA to reimburse physicians and other providers whether or not such providers are in their insurance plan’s network.
This is no different, however, than what is typically allowed under other health insurance plans. Individuals already can go outside their plan’s network (although they often have to pay higher cost-sharing if they do) under many existing, lower-deductible insurance plans, such as Preferred Provider Organization (PPO) plans or point-of-service HMOs that allow some out-of-network services. Since PPO plans comprise over 90 percent of the high-deductible plans attached to HSAs, this feature is increasingly common in health insurance arrangements regardless of whether HSAs are involved.
Moreover, any costs incurred by an individual with a HSA for these out-of-network services do not count toward meeting the high deductible. Only in-network services count toward the deductible or toward the overall limit on out-of-pocket costs that high-deductible plans attached to a HSA must carry. (This is the case with low-deductible plans as well.)
Under the law, the minimum high deductibles of $1,050 for individual coverage and $2,100 for family coverage — and the maximum out-of-pocket limit of $5,250 for individuals and $10,500 for family coverage — apply only to in-network services. There is no statutory limit for deductibles and no out-of-pocket limit for out-of-network services.
In short, individuals with HSAs face financial disincentives for out-of-network services similar to those faced by individuals in more comprehensive plans.
12. Question: Would patients have greater freedom to obtain medical services under HSAs by bypassing managed care and other health-insurance coverage restrictions?
Answer: Some proponents of HSAs claim that Health Savings Accounts free patients from onerous insurance rules that limit the use of key benefits or that deny coverage of such benefits entirely, such as the coverage restrictions imposed by managed care that provoked a backlash in the late 1990s. This is not correct. All health insurance plans today — including the high-deductible plans attached to HSAs — have coverage rules. Individuals typically must obtain prior approval from their insurer before certain benefits are covered. Other benefits may be limited in their amount, duration and scope, or not covered at all. While individuals can use tax-favored savings held in their HSA to pay for virtually any medical service on an out-of-pocket basis, the costs will not count toward meeting the high deductible that a plan sets unless the service would otherwise have been covered by the plan.
For example, if an individual did not get approval from the insurer for a service that requires prior authorization, the costs related to that service will not count toward either the deductible or the overall out-of-pocket limit under the plan. As a result, individuals with HSAs who have not yet met their high deductible likely face financial and administrative barriers in accessing services similar to barriers faced by individuals in more comprehensive plans.
Furthermore, once the high deductible has been met, people in these plans must comply with the same coverage rules as individuals in more comprehensive plans. Otherwise, their expenses are not covered.
13. Question: Would physicians and other providers receive higher reimbursements under HSAs then they do under existing health insurance plans because patients with HSAs would negotiate their own payment rates with providers?
Answer: Some physicians and other providers may support HSAs because they believe that individuals using funds held in their HSA will pay higher reimbursement rates — and in cash — than such providers would receive if they were reimbursed by an insurance plan. Individuals with HSAs, however, have an incentive to pay no more than the same discounted insurance reimbursement rates when using their HSA funds, because under the rules of the high-deductible plans, only the insurance plan’s regular reimbursement rate is likely to count toward the high deductible.
In other words, if an individual with a HSA pays a physician $250 for an office visit but the high-deductible plan will pay only $150 for that visit (after the high deductible is met), only $150 of the individual’s costs will be counted toward meeting the high deductible. Also, to the extent that an individual who has not yet met the high deductible has insufficient funds in his or her HSA to reimburse the provider for a medical service, the physician may receive less than if the service had been covered by the individual’s insurance plan, and the physician could receive nothing at all.
 The Joint Committee on Taxation has a somewhat lower cost estimate; it estimates the cost of the Administration’s HSA proposals at $108 billion over ten years. For an analysis of these proposals, see Jason Furman, “Expansion in HSA Tax Breaks Is Larger — and More Problematic — Than Previously Understood,” Center on Budget and Policy Priorities, Revised February 7, 2006 and Jonathan Gruber, “The Cost and Coverage Impact of the President’s Health Insurance Budget Proposals,” Center on Budget and Policy Priorities, February 15, 2006.
 For an analysis of why Health Savings Accounts raise these concerns, see Edwin Park and Robert Greenstein, “Latest Enrollment Data Still Fail to Dispel Concerns About Health Savings Accounts,” Center on Budget and Policy Priorities, Revised January 30, 2006.
 See Park and Greenstein, op cit and Edwin Park and Robert Greenstein, “Administration Defense of Health Savings Accounts Rests on Misleading Use of Statistics,” Center on Budget and Policy Priorities, February 16, 2006.
 America’s Health Insurance Plans (AHIP), “January 2006 Census Shows 3.2 Million People Covered by HSA Plans,” March 9, 2006.
 General Accountability Office, “Consumer-Directed Health Plans: Small but Growing Enrollment Fueled by Rising Cost of Health Care Coverage,” GAO-06-514, April 2006.
 GAO, op cit.
 Mercer Human Resource Consulting, “Beyond the Early Adopters: Consumerism at Work in the Marketplace,” February 6, 2006.
 Gary Claxton, Jon Gabel, Isadora Gil et al., “What High-Deductible Plans Look Like: Findings from a National Survey of Employers, 2005,” Health Affairs, Web Exclusive, September 14, 2005.
 Claxton et al., op cit.
 IRS guidance indicates that some prescription drugs may meet the preventive care definition, such as cholesterol-lowering drugs taken by an individual without heart disease to prevent the future occurrence of heart disease (although it is unclear to what extent high-deductible health insurance plans actually cover such drugs), but the preventive care exemption does not encompass prescription drugs used to treat an existing illness, injury, or condition, including conditions that could worsen in the absence of such medications. See Internal Revenue Service, “Internal Revenue Bulletin: 2004-33,” August 16, 2004. See also Edwin Park, “Health Savings Accounts Unlikely to Significantly Reduce Health Care Spending,” Center on Budget and Policy Priorities, June 12, 2006.
 McKinsey & Company, “Consumer-Directed Health Plan Report — Early Evidence Is Promising,” June 2005. A HRA differs from a HSA in that it is employer-funded, not portable, and does not have the same tax advantages as a HSA.
 Paul Fronstin and Sara Collins, “Early Experience with High-Deductible and Consumer-Driven Health Plans: Findings from the EBRI/Commonwealth Fund Consumerism in Health Care Survey,” Employee Benefit Research Institute, December 2005.
 For a summary of the tax benefits of HSAs and why they are likely to be used as tax shelters, see Park and Greenstein, “Latest Enrollment Data Still Fail to Dispel Concerns About Health Savings Accounts.”
 See Furman, op cit.
 GAO, op cit.
 For an examination of why enrollment data from the individual market may not be particularly useful, see Park and Greenstein, “Latest Enrollment Data Still Fail to Dispel Concerns About Health Savings Accounts.”
 eHealthInsurance, “Health Savings Accounts: January 2005 — December 2005,” May 10, 2006.
 Leonard Burman, “Medical Savings Accounts and Adverse Selection,” Urban Institute, 1997.
 Burman, op cit and David Cutler and Sarah Reber, “Paying for Health Insurance: The Tradeoff Between Competition and Adverse Selection,” The Quarterly Journal of Economics, May 1998.
 The LASIK market was found to suffer from deceptive advertising and other consumer protection problems.
 See Paul Ginsburg, Written Testimony before the Subcommittee on Health, U.S. House Committee on Energy and Finance, “Consumer Price Shopping in Health Care,” Center for Studying Health System Change, March 15, 2006 and Paul Ginsburg, Statement before the Joint Economic Committee, Center for Studying Health System Change, May 10, 2006. More in-depth working papers on this subject are available upon request from the Center for Studying Health System Change.
 Mercer Human Resource Consulting, “Consumer-directed health plans were least costly plan in 2004,” February 21, 2005. In some cases, HSA-qualified high-deductible plans may be more expensive than plans not qualified for HSAs with similar high deductibles. See Sarah Rubenstein, “High-Deductible Shoppers Find HSA Plans May Cost More,” Wall Street Journal, June 6, 2006.
 Deloitte Consulting LLP and the Deloitte Center for Health Solutions, “Reducing Corporate Health Care Costs 2006 Survey,” February 2006.
 Beth Fuchs and Julia James, “Health Savings Accounts: The Fundamentals,” National Health Policy Forum, April 11, 2005.
 eHealthInsurance, op cit.
 See AHIP, op cit and America’s Health Insurance Plans, “Number of HSA Plans Exceeded One Million in March 2005,” May 18, 2005. In the individual market, while the average annual high deductible for the best-selling HSA-eligible high deductible insurance plan declined by 17 percent for individuals and 10 percent for family coverage between 2005 and 2006, the average out-of-pocket limit increased by 18 percent for individuals and 27 percent for family coverage. In the small group market, the average annual deductible for the best-selling HSA-eligible plan increased by nearly 16 percent for individuals and nearly 8 percent for family coverage, and the average out-of-pocket limit increased by 4 percent for individuals and declined by 1 percent for family coverage. In the large group market, the average annual deductible for the best-selling HSA-eligible plan increased by 9 percent for individuals and nearly 17 percent for family coverage, and the average out-of-pocket limit increased by 4 percent for individuals and less than 1 percent for family coverage.
 Fuchs and James, op cit.
 Fuchs and James, op cit and Timothy Jost and Mark Hall, “The Role of State Regulation in Consumer-Driven Health Care,” American journal of Law and Medicine, v.31 (2005), p.395-418
 AHIP (2006), op cit.
 Fuchs and James, op cit, and Jost and Hall, op cit.
 Fuchs and James, op cit, and Jost and Hall, op cit. The one difference is that out-of-pocket costs related to such out-of-network services would receive a tax subsidy if paid for with funds held in a HSA.
 Fuchs and James, op cit, and Jost and Hall, op cit.
 Fuchs and James, op cit, and Jost and Hall, op cit.