Extending U.S. Medicare to Mexico

This paper explores the idea of extending U.S. Medicare benefits to Mexico—that is, allowing Medicare-eligible beneficiaries to use Medicare to cover health services received in Mexico. We provide an overview of both the Medicare program and the Mexican health care system in their current state; discuss the potential number of Medicare-eligible beneficiaries residing in Mexico today and over the next several decades; review why extending Medicare to Mexico is important to consider from the perspectives of Medicare-eligible consumers, the U.S. government, and the Mexican government; offer a series of potential policy options; and discuss the feasibility of these options, along with logical next steps.

Our population of interest is any Medicare-eligible beneficiary who might receive health services in Mexico. For the purposes of this paper, we define a Medicare-eligible beneficiary as someone age 65 or older who has worked (or whose spouse has worked) at least 40 quarters in Medicare-covered employment in the United States. Such persons are eligible to enroll in Medicare Part A (Hospital Insurance [HI]) for free and Medicare Part B (Supplementary Medical Insurance [SMI]) for a monthly premium and are not required to maintain residency in the United States. Others qualifying for Medicare, namely persons with disabilities or end-stage renal disease, or persons not eligible for free HI through their employment but who can pay for HI if they meet a U.S. residence test, are not included in this consideration.

Medicare's Lack of Coverage Abroad

The U.S. Social Security Medicare Program does not cover health services outside of the United States.[1] Medicare-eligible beneficiaries who travel or reside abroad must obtain some form of supplemental insurance to cover health care costs incurred abroad or pay out-of-pocket. The policy may be a disincentive for retirees in the United States to travel or live abroad and certainly presents an inconvenience for those who do travel or live abroad. On the surface, this policy may be viewed as a potential cost-saver for Medicare: Medicare is not required to reimburse costs incurred abroad nor to develop what could certainly be a costly administrative infrastructure to certify providers and reimburse for care abroad.

Many Medicare beneficiaries purchase Medigap (or “Medicare Supplemental Insurance”) policies to cover gaps in their Original Medicare (Parts A and B) coverage, such as coinsurance, co-payments, and deductibles. Medigap policies are standardized private health insurance policies designed to supplement Original Medicare. Of the standard Medigap policies A through L, C through J (or eight of twelve) offer a Foreign Travel Emergency Benefit up to plan limits. However, the benefit applies only when beneficiaries are away from the United States for six weeks or less, covers emergency services only, and requires the beneficiary to pay a separate deductible of $250 USD per year.[2]

Other Medicare beneficiaries choose a Medicare Advantage Plan in lieu of Original Medicare to reduce out-of-pocket health care costs. These plans tend to cover more services than Original Medicare and have lower co-payments, but beneficiaries may be restricted to health care providers and facilities that are part of the plan. Medigap policies do not work under Medicare Advantage Plans, so beneficiaries choose one or the other. In theory, Medicare Advantage Plans should have the ability to offer coverage abroad as a supplemental benefit, but, in practice, they do not, with the possible exception of optional coverage for travel emergencies. Lack of coverage abroad within Medicare Advantage may be attributed to actual or perceived Medicare regulatory or legal restrictions, or to a lack of demand in the market.

Other federal health benefit programs, available to specific groups of retirees, do offer coverage abroad. For example, military retirees, their families, and survivors receive TRICARE for Life benefits when they become eligible for Medicare at age 65 and purchase Part B. TRICARE for Life acts as a “wraparound” policy to Medicare, filling in gaps in Medicare coverage. It covers care abroad, with beneficiaries being responsible for TRICARE Standard cost shares and deductibles.[3] In Mexico, beneficiaries can visit any licensed host-nation provider for care. TRICARE assists beneficiaries in finding qualified or preferred overseas providers, both through its own service center as well as through its partner, International SOS—the world's largest medical and security assistance company. In most cases, beneficiaries have to pay the provider up front and file a claim with TRICARE for reimbursement.[4]

Similarly, U.S. federal employees continue to receive the same health benefits in retirement that they enjoyed as employees, and all Federal Employees Health Benefit (FEHB) plans offer international coverage for their beneficiaries traveling or living abroad.[5] Other non-health-related federal benefits, such as Social Security retirement benefits, do not require U.S. residency as a condition for benefit claims. In addition, U.S. citizens living abroad are required to continue paying U.S. income tax, and the recently passed U.S. health reform legislation includes a new 3.8 percent Medicare tax on investment income beginning in 2013 for individuals earning more than $200,000 USD annually ($250,000 USD for couples). This new tax will significantly affect retirees, who typically rely on investment income for their living expenses. These examples raise valid questions as to why Medicare coverage, or some equivalent alternative, is not being extended to services rendered outside of the United States—particularly for those residing abroad.

Why Is This Particularly Relevant to Mexico?

Although it is difficult to establish the number of Medicare-eligible retirees in Mexico, the number is not insignificant and is rising. There are primarily two different groups of interest: Americans who have decided to retire in Mexico (part- or full-time) and Mexicans who have returned to Mexico in retirement after living and working in the United States as U.S. permanent residents. However, available data on both groups are lacking. The U.S. Department of State used to issue estimates of the number of Americans living abroad by country, but ceased issuing these numbers due to security concerns. The U.S. Government Accountability Office (GAO) examined the feasibility of counting Americans overseas as part of the U.S. Census but, following a pilot test in 2004, determined that this was not cost-effective.[6] The U.S. Embassy in Mexico has access to data on Americans who register with the embassy, but registering is voluntary and numbers are thought to be gross underestimates.

The best available data sources for estimating the number of Medicare-eligible retirees in Mexico are the Mexican Census and the U.S. Social Security Administration (SSA). Both, however, provide underestimates of the number of Medicare-eligible persons in Mexico. Data from the 2000 Mexican Census indicate 17,554 Americans age 65 or older residing in Mexico.[7] But because the census defines nationality by country of birth, not citizenship, this number does not include U.S. citizens born outside of the United States (Mexican or other). It also does not include Mexicans who were formerly U.S. permanent residents. In addition, the census data are unlikely to capture many American retirees who reside in Mexico only part-time.

In 2008, the SSA reported that there were 23,583 retired social security beneficiaries living in Mexico.[8] While this number is more recent and should include former U.S. permanent residents who worked in the formal sector in the United States, it is also likely to be a significant underestimate due to many retirees maintaining residences in the United States and keeping that address on file with SSA. Those who have conducted extensive related research and fieldwork in Mexico, particularly in American expatriate communities, assert that the number of American retirees currently living in Mexico (full- or part-time) is well above 100,000.[9] Adding in the lesser known, but substantial, number of Medicare-eligible Mexicans returning to Mexico after years of living and working in the United States, the numbers take on increasing significance for the U.S. Centers for Medicare and Medicaid Services (CMS).

Despite the lack of hard data, two points regarding U.S. migration to Mexico are not in dispute. First, Mexico is home to a very large number of Americans living outside of the United States. In 1999, the Department of State estimated that, of the 6.6 million Americans living abroad, 1.04 million were residing in Mexico.[10] Of the total population of retired American workers (U.S. citizens and permanent residents receiving social security retirement benefits) living abroad, Mexico is home to the second-largest group (Canada being the first).[11] Meanwhile, 2000 Mexican Census data reported by Mexico's National Population Council (CONAPO) indicate that migrants from the United States make up the largest portion, by far, of the country's foreign-born population.[12]

Second, the number of Americans, particularly retirees, living in Mexico is increasing rapidly and is likely to continue to increase in the coming decades. Mexican Census data from 2000 published by CONAPO show a 76 percent increase in the number of U.S. natives living in Mexico between 1990 and 2000, and a 45 percent increase among those age 65 years or older.[13] Importantly, the increase was significantly higher in municipalities with established expatriate retiree communities (e.g., 581 percent in Chapala).[14]

Assuming no change in the rate of increase, the number of U.S.-born retirees living in Mexico could more than triple by 2030. However, as the baby boom generation is just on the cusp of retirement, the number of U.S. retirees will begin to grow at unprecedented rates. The U.S. Census Bureau estimates that the U.S. population age 65 or older will more than double between 2000 and 2040 and that, although the rate of growth will slow over time (from approximately 16–17 percent every five years between now and 2025 to 4–5 percent every five years between 2035 and 2050), the number of retirees will still be increasing in 2040 and beyond.[15] As the number of U.S. retirees increases, the numbers emigrating to Mexico seem certain to increase. This can be attributed to the “pull” of bargain prices south of the border (on real estate, property taxes, and health care, to name a few), ideal climates, and growing networks of Americans in Mexico, as well as to “push” factors such as climbing health care costs, shrinking pensions, and the perception of a declining quality of life in the United States.[16]

It is also important to consider the growth rates of Medicare-eligible return migrants to Mexico. CONAPO projects that net immigration to the United States will continue at between 3.5–5 million per decade through at least 2030.[17] Long-term migrants in the United States have been found to be particularly vulnerable to a lack of public health insurance coverage upon return to Mexico, compared with short-term migrants and non-migrants.[18]

How Are Medicare-Eligible Beneficiaries in Mexico Seeking Health Care Today?

Health care in Mexico is delivered by a range of different institutions with few connections among them. About half of the population is insured through their relevant social insurance institution—the Mexican Institute of Social Security (IMSS) for salaried workers in the formal sector, the Institute of Security and Social Services for Government Employees (ISSSTE), and several others for employees of the military, navy, and Petroleum of Mexico (PEMEX). Each of these institutions employs salaried doctors and owns and operates its own hospitals and clinics. The social security systems are currently financed by a combination of employee and employer contributions, along with transfers from the federal government. The so-called “uninsured” population (i.e., unemployed, self-employed, or employed in the informal sector), although not covered by an insurance institution, can still access health care at significantly lower cost than full-market prices in publicly financed Ministry of Health (MoH) and state facilities. In addition, the MoH has launched a program to extend insurance coverage to those without social security. This Popular Health Insurance, or Seguro Poplar de Salud (SP), provides families with access to health insurance on a voluntary basis. Finally, private insurance exists, but it covers less than 3 percent of the population—often individuals already covered in the social security system.[19]

While Mexican patients appear to be largely satisfied with the public care they receive, a large share of private spending across all health care expenditures suggests problems related to quality of care and system efficiency and responsiveness. Lack of capacity in the public sector has led to demand spilling over into the private sector, which is large but mostly unregulated, and for which there is little insurance coverage.[20] There are reports of wide variability in quality, however, both across and within the public and private sectors.[21] As part of an overall vision of the Mexican government, the National Crusade for Quality was introduced under the National Health Programme (2001–2006) to measure and improve the quality of health services across Mexico.

American and former U.S. permanent resident retirees living in Mexico have a variety of options to pay for health care services. Within the Mexican health care system, they can “buy into” IMSS for a modest annual premium,[22] pay out-of-pocket for private health services, purchase a private insurance plan, or seek care at MoH or state facilities by either paying out-of-pocket or purchasing SP insurance. Of these four options, buying into IMSS and paying out-of-pocket for private health care are most prominent. MoH and state facilities are primarily used by the poor, and comprehensive private insurance, which is underdeveloped in Mexico, has, at least in the recent past, been unjustifiably expensive given other available options. The fifth option is to maintain Medicare and travel back to the United States for health care services. It is important to note, however, that this option can easily become unavailable to former U.S. permanent residents, as Lawful Permanent Residents living abroad for longer than six months lose their residency status.[23] Thus, individuals who would like to return to the United States to receive Medicare benefits may not be allowed to enter the country without the often time-consuming process of obtaining a visa.

Our understanding is that American and former American resident retirees living in Mexico use a combination of these options, based on personal circumstances and service needs. Many do not pay Medicare premiums (retaining only Part A, which is free) and instead use a combination of IMSS and private services in Mexico for all of their health care needs, but they know that they can return to the United States for hospital care if required. Others maintain Part B, as well as other Medicare-related products (e.g., Part D [prescription drug coverage], Medigap policy) in case they decide to return to the United States for medical care. Although many buy into IMSS, those with the financial means may not use it exclusively. That is, they may use the private sector to enhance their provider choices and avoid wait times.

Why Consider Extending Medicare to Mexico?

The notion of extending Medicare to Mexico is sure to meet with opposition, or, at the very least, a great deal of skepticism. Concerns will be raised about financing the program, ensuring the quality of care, establishing a means for financial oversight, and justifying extension to Mexico while continuing to exclude coverage in other countries. In thinking about the possible extension of Medicare to Mexico, it is important to consider the arguments for such an extension from each of the relevant perspectives. This section briefly covers how an extension of Medicare to Mexico could benefit various relevant parties, providing justification for considering the proposition.

The Consumer Perspective: Americans and Former American Permanent Residents Living in Mexico

Medicare-eligible beneficiaries living in Mexico, of course, would like to be able to receive their Medicare benefits in Mexico. After investing in the system for years, they receive no benefit—direct or substitute—in the communities in which they live. American citizens living in Mexico do maintain access to Medicare benefits in the United States, but this requires often significant travel expenses and is sometimes very difficult or impossible due to health conditions. For Medicare-eligible noncitizens, access is more limited, as U.S. residency expires after six months of living abroad, and re-entering the U.S. requires what can be a lengthy process of obtaining a visa.

The American expatriate retirement community in Mexico is quite vocal about its desire to see Medicare extended to eligible beneficiaries abroad and has begun lobbying for consideration of such an extension.[24] However, the group seems to be pushing for change without being fully informed or thinking through the potential consequences. For instance, many believe that Medicare coverage would reduce their out-of-pocket costs when, in reality, Medicare Part B premiums (at least currently, for care obtained in the United States) are much more expensive than buying into IMSS and, in many cases, are more expensive than paying out-of-pocket for private health care services in Mexico.

Aside from the issues of fairness and cost, not providing Medicare benefits to eligible retirees also could have negative implications for the quality of care that Medicare-eligible beneficiaries residing in Mexico are receiving. Presumably, retirees are obtaining health care services from a wide variety of systems and providers both within Mexico and across the United States. The use of multiple providers, particularly across different countries with different languages, reduces the continuity of care, limiting the physician-patient relationship and creating problems with respect to the flow of health information. These elements of quality of care are particularly important in the elderly population, which is characterized by complex chronic health conditions. Thus, even within a system such as Mexico's, in which there are concerns about variability in quality of care, Medicare-eligible beneficiaries might receive better care, overall, by remaining in Mexico because of improved continuity of care and better information flows.

The U.S. Government Perspective

Medicare is facing very severe financial difficulties—worse than the more publicized financial woes being confronted by Social Security. Health care costs are rising faster than the earnings per worker, on which payroll taxes and Medicare benefits are based. Current projections indicate that expenditures will outstrip revenues by 2012, and that by 2017, HI Trust Fund assets will be exhausted—two years earlier than projected in 2008.[25] There is a great deal of pressure in Washington to expand Medicare revenues or limit Medicare expenditures, and controlling the rapid increase in health care costs must be part of the solution.

To the extent that individuals are traveling back to the United States to receive health care rather than consuming less-expensive care in Mexico, expanding Medicare to Mexico may reduce Medicare expenditures. This, of course, would also be dependent on the structure of the extended benefit and its associated costs. Extending Medicare to Mexico also has potential implications for quality improvement. If Medicare-eligible beneficiaries stay in Mexico and obtain their care within one health care system (e.g., IMSS) rather than bouncing through a mixture of both Mexican and American health providers, there may be improved chances that Medicare beneficiaries will receive the right care at the right time[26]—another potential avenue for reducing Medicare expenditures over time. Receiving the right care at the right time, however, will depend in part on the overall quality of the Mexican health care system, the quality of its practitioners, and access to a wide range of medical tests and equipment.

The Mexican Perspective

Although little is known about the health-care-seeking behaviors of Medicare-eligible retirees living in Mexico, we know that many buy into IMSS as a source of health care coverage. The coverage is comprehensive and the cost modest. However, Mexican officials are concerned about the sustainability of the buy-in program as the number of foreigners and return migrants in Mexico increases. IMSS was designed primarily to support Mexican taxpayers who pay into the system for decades, and there are concerns about it being “overrun” by foreigners if the number of migrants continues to increase as expected.[27] These problems could be addressed by finding the buy-in “price” that both covers the costs imposed by the Medicare-eligible enrollees and prevents demand from outstripping supply.

Expanding Medicare to Mexico would presumably alleviate the potential drain on IMSS, but it could also have significantly more important, broader effects, such as improving health care quality, attracting additional Americans to Mexico, and expanding Mexico's health care economy by increasing the amount of care provided. Any scheme for the portability of Medicare benefits would likely involve some level of oversight or additional regulatory requirements, which could contribute to enhanced quality and consistency of care within the Mexican health care system. In addition, a study examining U.S. retirement migration to Mexico suggests that Medicare portability might have strong effects on the number of U.S. retirees moving abroad.[28] With an increased number of current residents remaining in Mexico for care comes an expanded health care market, as well as other sorts of benefits for Mexico. American retirees in Mexico contribute to civic society in their local communities (e.g., charitable and volunteer work), bring an influx of capital, and contribute to job creation.[29]

What Are the Options for Medicare Extension?

There are several options that could be considered for extending Medicare coverage to Mexico (or elsewhere). Here we describe four policy options (referring to them as Medicare Abroad) and the issues that would need to be addressed in order to purse each.

Option 1: Expand Original Medicare to Allow Reimbursement of Mexican Providers

The Original Medicare fee-for-service program consists of Part A and Part B, which cover inpatient and physician services, respectively. Medicare-eligible participating providers are reimbursed directly for their services, with reimbursement rates determined by a complex set of rules. In theory, the administrative infrastructure could be expanded to allow Mexican providers to be reimbursed through Parts A and B, thereby expanding Medicare coverage to Mexico. Because of the complex rules governing reimbursements, and the differences in the health economies of Mexico and the United States, such an expansion would require a substantial investment in administrative infrastructure to address the determination of (1) Medicare-eligible provider status, (2) reimbursement rates, (3) means for coordination with existing Mexican institutions, and (4) premium rates.

Determination of Medicare-eligible provider status. Under the current Medicare rules, Mexican providers would have to be certified as Medicare-eligible to qualify for reimbursements. This would entail the establishment of minimum quality criteria, as well as a willingness to participate among Mexican providers. Certification might be achieved in collaboration with Mexican institutions, such as the MoH, and would accomplish several things. First, it would enable Medicare to require providers to report cost and quality data (for example, the submission of detailed hospital cost data by department, by which diagnosis related group [DRG] reimbursement rates are determined). It also would provide a means to establish the authenticity of providers, limiting the potential for fraud and abuse. Finally, it would provide a means of enabling oversight by which quality and efficiency could be improved over time. However, unless there is some expectation that Medicare patient volume will be great enough, Mexican institutional health care providers may not be willing to participate in a costly Medicare certification process.

Determination of reimbursement rates for Mexican providers. There are roughly 20 categories of service by which Medicare pays providers. Each has its own rules stipulating schedules for payment, and each would have to be adapted to deal with the differences between the Mexican and U.S. health economies. For example, Medicare pays physicians based on analyses of the resources used for each type of encounter, referred to as a resource-based relative value scale (RBRVS). Inpatient admissions are reimbursed for bundles of services based on the DRG of the admission. Reimbursement rates for DRGs are based on national hospital cost accounting, with very little geographic variation. In addition, Medicare requires that these reimbursements be accepted as payment in full and, except for applicable co-payments, prohibits balance billing, whereby providers collect any remaining charges directly from the Medicare beneficiary. Given differences in patterns of care, resource use, and prices in Mexico, each of the reimbursement schedules (and methodologies) would have to be revisited. Because there are more than 40 million Medicare enrollees in the United States, the complex infrastructure generates a very small administrative cost per enrollee. However, with significantly fewer potential beneficiaries to share the costs, duplication (or modification) of this infrastructure for Mexico could be prohibitively expensive. In addition, operation of this kind of Medicare program would likely require close oversight and regulation of Mexican medical institutions by a U.S. regulator (CMS).

Coordination with existing Mexican institutions. In the United States, Medicare is always the first payer when an individual has additional types of insurance. In these cases, the additional insurance (e.g. Medicaid, Medigap, or other supplemental insurance) may cover part or all of the Medicare co-payments and deductibles for which the Medicare beneficiary would otherwise be responsible. With the expansion of Medicare Parts A and B to Mexico, structures would be required to coordinate coverage from the (potentially) multiple insurers. For example, Mexico, which currently allows Americans to buy into IMSS for a modest price, would need to revisit this policy to determine the best way to coordinate coverage with Medicare, perhaps preventing Medicare-eligible residents from purchasing IMSS. At a minimum, appropriate coordination of coverage would require administrative oversight, which could be provided by the Mexican government, CMS, or both.

Determination of premium rates. The additional administrative costs that would be required to establish Original Medicare in Mexico would be substantial. Enrollment could require the payment of additional premiums by Mexican residents to cover these costs, but they would likely be considerable. The current standard monthly Part B premium is $110.50 USD, which already substantially exceeds the maximum annual IMSS buy-in price. Additional premiums required to cover administrative costs would make the Original Medicare program in Mexico particularly unattractive, especially when compared against the costs of buying into IMSS. Mexico, of course, could consider preventing Medicare-eligible individuals from joining IMSS, but expecting Mexican residents to pay such a high premium might still be unrealistic given the relatively low cost of paying out-of-pocket for care in Mexico. Alternatively, coverage in Mexico could be automatically included as part of all Original Medicare enrollment, with the additional administrative costs borne by all Medicare enrollees. This would, of course, eliminate the disincentive to enroll, and it could be financed in part by reductions in expenditures through the utilization of lower-priced care in Mexico.

Option 2: Develop a New Program to Cover Services Abroad

A second policy option would be for Medicare to develop a traditional indemnity insurance plan that would cover some or all of the services covered by Original Medicare. Such a plan (potentially Medicare Part E) would serve as a third-party payer, with CMS making fee-for-service payments directly to providers or to beneficiaries as reimbursement for medical services. It could be designed to mirror other insurance programs with international coverage, such as TRICARE for Life, taking advantage of their existing administrative infrastructures.

Although many of the same issues described in Option 1 would still require consideration, the most burdensome would be avoided: Medicare would reimburse based on Mexico's existing price system in lieu of developing a new, complex reimbursement rate infrastructure specific to Mexico. Because this is a Medicare program funded by taxpayer dollars, some—and perhaps a substantial—level of oversight might be required to address concerns about the potential for fraud and abuse, as well as about quality and quality improvement. In addition, coordination with the Mexican MoH and other relevant institutions would be necessary to address concerns about multiple insurance, additional elements of fraud, means to certify quality, and mechanisms to improve quality. Such collaboration could significantly reduce the costs of what would otherwise be duplicative Medicare administrative structures. At the same time, it could provide opportunities for capacity-building and quality improvement within Mexican institutions, as well as contribute to the development of stronger ties and working relationships between the U.S. and Mexican health bureaucracies.

Option 3: Expand Medicare Advantage

Medicare Advantage plans are managed care organizations that enroll Medicare-eligible individuals and receive capitation payments from Medicare in return for the provision of patient care. Expansion of Medicare Advantage could be a relatively easy way to expand Medicare coverage to beneficiaries living abroad. This would entail relatively little additional administrative costs for CMS, as the Advantage plans would be responsible for administering the provision of and reimbursement for covered services provided in Mexico. Unlike Option 2, the market would determine the extent to which the Advantage plans are available, and the Advantage plans, rather than CMS, would bear the cost risk.

This option could lead to the development of niche plans that specialize in services abroad. By relying on the market, it could result in the development of plans and coverage that best meet the needs and interests of enrollees. The development of such plans, however, might be limited if the demand is not sufficient. While this might not be a problem if large Advantage Abroad plans developed with operations across several different counties, the substantial differences in health care systems across countries could limit the potential for such multi-country plans. Country-specific plans (e.g., a Medicare Advantage Mexico) would have to have high enough enrollments to sustain the costs required for administration. Given the varying estimates of the current number of Medicare-eligible individuals living in Mexico, this is currently an open question. Such programs may be more viable by 2040, with substantial increases in the number of potential enrollees.

Option 4: Expand Medigap

Many of the standard Medigap plans (eight of twelve) already provide limited insurance against medical costs for beneficiaries traveling abroad, through their Foreign Travel Emergency Benefit option. Unlike Options 1 through 3, Medigap expansion could be implemented in whole or in part through incremental changes. Incremental expansions could extend coverage to longer stays abroad (beyond the current 60-day limit), eventually including permanent residents abroad, or they could, over time, be expanded beyond the current emergency coverage to include all health care. Because the model already exists and extensions in either dimension could be introduced incrementally, expanding Medigap would pose fewer hurdles than the other options. As the program grows, it may become necessary or desirable to implement oversight and cost-containment policies similar to those described under Option 1, including the determination of eligible providers and reimbursement rates. Because incremental expansion will be more limited (and predictable) in size, these expansions initially would have only a small effect on the Mexican health care economy and Medicare expenditures. As it grows over time, however, incremental expansion could have the potential to realize the same economic and health care quality improvement advantages as those described for the other options.

From the perspective of the beneficiary, Option 4 could evolve to look very similar to Option 2. From the perspective of CMS, however, the two options are quite different. With Option 4, Medigap plans would bear the cost risk and would have to collect a premium, separate from Part B, to cover the costs. These premiums, presumably, would be paid at least in part by CMS as a reflection of the shift in medical costs from CMS to the Medigap plan.

How Feasible Are These Options?

The four policy options presented above differ greatly in their practical and political feasibility. In addition, while these options could be adopted in combination, the relatively thin market in Mexico makes this unlikely, as the duplication of administrative costs would likely prove prohibitive, and the success of at least one option (i.e., Option 3 to expand Medicare Advantage) would rely on market share.

Practical Feasibility

In the U.S. economic and political environment of the foreseeable future, the adoption of an extension to Medicare—which could be perceived as increasing overall Medicare program costs and reducing U.S. employment—would be difficult to achieve. There are several features of the Mexican and U.S. health economies, however, that suggest that a policy to extend Medicare to Mexico might be beneficial to all parties. American taxpayers might benefit from a reduced total cost of Medicare: To the extent that an extension of Medicare, through any of the above options, induces Medicare beneficiaries to substitute higher-cost U.S. health care services with lower-cost Mexican services, overall Medicare expenditures might be reduced if the extended benefit is suitably structured. In addition, beneficiaries would likely be better off because of improved access to local care and reductions in the pecuniary and time costs of travel to the United States for care. Finally, Mexicans might benefit from an expanded health economy and improved quality of care. The potential for a mutually beneficial outcome certainly justifies further consideration of extending Medicare to Mexico.

Although the numbers are questionable and estimates variable, the size of the Medicare-eligible population in Mexico will never reach a threshold justifying the per person costs of administering Option 1, particularly given the other options currently available to Medicare-eligible beneficiaries living in Mexico (i.e., buying into IMSS or paying out-of-pocket). Option 2, with reimbursement rates set at some fraction of the U.S.-based payment, or set to cover some fraction (perhaps 100 percent) of charges, would still require an administrative infrastructure to address issues of fraud and abuse, certification and quality, and coordination across multiple insurers. It could, however, be developed and implemented in coordination with existing programs for Americans abroad, including TRICARE for Life and FEHB plans, as well as Mexican institutions, such as the Mexican MoH. Cost considerations strongly favor Option 2 over Option 1 for a full-scale Medicare Abroad program.

The practical feasibility of Option 3 (expansion of Medicare Advantage) would depend on whether such insurance plans could be supported by the market. In Mexico, the costs of administrating such a plan (borne by the private insurer) could be prohibitive given the number of Medicare-eligible retirees in Mexico. Although both the current Medicare requirement that services be provided within the United States and the relatively small population of potential enrollees residing in Mexico may explain the nonexistence of these programs today, a relaxing of the requirement along with the anticipated substantial growth of potential enrollees by 2040 could make these programs viable.

Option 4 is perhaps the most practically feasible, as it requires no new program and can be implemented incrementally. By rolling out reform incrementally, initial investments in the development and operation of administrative infrastructures could be modest. In addition, the program could be shaped to address the most important needs of beneficiaries as it is rolled out, addressing the highest-priority areas with each successive incremental expansion.

Option 1 is economically and practically not viable given the expected administrative costs and the relatively small population affected. Given its reliance on the market, Option 3 is also unlikely to be viable even with the anticipated large population increase among potential enrollees. Multiple Medicare Advantage Plans competing for enrollees would result in very high administrative costs that would have to be borne by plan enrollees. Options 2 and 4—developing a new Medicare indemnity insurance plan to cover services abroad or expanding Medigap—seem the most viable. Over time, Options 2 and 4 could eventually look very similar. Option 4, however, has the practical advantage of allowing for incremental expansion, minimizing both practical and political risks. For example, immediate incremental changes may require only changes to Medicare regulation, while more fundamental reforms under each of the four options would require statutory changes through an act of Congress.

Political Feasibility

Although a Medicare Abroad program in Mexico, based on any of the above options, might benefit Medicare-eligible beneficiaries, the U.S. government, and Mexico, there are three substantial political issues that would have to be overcome: (1) a call for parity across other countries; (2) the perception of unfairness among U.S. resident American taxpayers, who might not feel they are equally benefiting; and (3) the perception of U.S. jobs being lost to Mexico among both health professionals and the general population.

Part of the justification for considering an extension of Medicare could come from Mexico's participation in the North American Free Trade Agreement (NAFTA). However, there is likely a much stronger economic case to be made for Medicare extension to Mexico than to Canada or some other countries enjoying free trade agreements with the United States (e.g., Israel, Australia, Korea, Bahrain). Would it be politically viable for Medicare to adopt dissimilar policies for different countries, particularly within NAFTA? Americans residing in Canada (or other countries) might raise legitimate concerns about equity. Similarly, the Canadian government might raise concerns about the burdens Americans impose on the Canadian health care system.

Another set of problems could arise around U.S. resident taxpayers articulating their distaste for Medicare funds being directed at programs that likely never will benefit them. Public relations, including communications on the potential cost savings for Medicare and responding to reports of fraud and abuse, would need to be emphasized. Even if fraud and abuse cases are limited and amount to a relatively small fraction of program costs, they could generate considerable political pressure. Thus, significant oversight and auditing would likely be required in any of the options, adding to administrative costs.

Finally, the magnitude of the resultant decrease in demand for Medicare-related services in the United States as a result of extending Medicare to Mexico will not likely be great enough to significantly affect any given individual physician, physician group, or institutional provider. There may be a perception, however, that extending Medicare to Mexico would reduce demand for services—and thus related employment—in the United States, generating opposition from physician groups, such as the American Medical Association, and the general population.

What Are the Steps Forward?

Given the possibility that an extension of Medicare to Mexico could benefit consumers, the Mexican government, and the American government, further consideration of the policy options and a detailed analysis of the issues make sense. Although we argue that Options 2 and 4 are likely the most sensible to pursue, we suggest modeling each option to assess the effects on various parties across several dimensions. A critical starting point for such a modeling exercise is to develop more-accurate estimates and projections for the size of the Medicare-eligible population in Mexico. Today's estimates vary significantly (from about 20,000 to 100,000), as do projections of future increases. A model could then be developed to demonstrate explicit benefits and costs of extending Medicare to Mexico under different policy scenarios. Such a model would assimilate information about population sizes, relative prices, behavioral responses, and program costs to generate estimates of the fiscal consequences for CMS, the quality of care in Mexico, and the expansion of the health economy in Mexico. We have included a short description of what such a modeling exercise would entail in Box A.

Box A: Medicare Abroad Policy Model

To estimate the effects of various policy options, a model could be developed that would have, at its core, an accounting of the number of Medicare-eligible Americans and Mexicans living in Mexico, their use of medical services (both in Mexico and the United States), and changes in their behavior that would be induced by alternative policies. The conceptual model diagram below shows, from left to right, these model inputs and outputs. The model would include estimates of how increased demand for care by Medicare-eligible individuals in Mexico, combined with Medicare certification requirements, could induce changes in the quality and price of care in Mexico. In addition, the model would include estimates of changes in satisfaction with care and how they could induce additional utilization of care in Mexico relative to the United States among Medicare-eligible individuals.

The model would require detailed data regarding the numbers of individuals, their utilization and location of care, and the cost of care (including the source of funds) in order to assess the following: (1) the total change in Medicare expenditures, (2) the total change in health care spending in Mexico, and (3) the total out-of-pocket expenditures of Medicare-eligible individuals in Mexico (including the costs of co-payments and deductibles, Medicare and/or IMSS premiums, and travel for care). These estimates, together with estimates of satisfaction with care and changes in the quality of care, would allow us to determine the net benefits of each policy change for each of the key groups of players (CMS, U.S. taxpayers, Medicare-eligible individuals living in Mexico, Medicare-eligible individuals living in the United States, and the Mexican government).

With information generated from use of the model, a strategy could be formed regarding which of the policy options, if any, to pursue, and the best combination of details within that option. Additionally, empirical evidence on the consequences of the policy could be gathered through the implementation of a Medicare demonstration project. Medicare demonstrations are commonly used by CMS in the exploration of potential policy changes, addressing the feasibility of proposed policies and the uncertainty of its consequences by adopting them on a limited scale. Medicare has performed demonstrations, for example, to examine care management for high-cost beneficiaries and the use of electronic health records. Although contained, demonstrations can be quite large, often including populations of beneficiaries in excess of even the highest estimates of the number of Medicare-eligible beneficiaries residing in Mexico today. Extending Medicare coverage to Mexico might be an ideal application for a Medicare demonstration, and the Association of American Residents Overseas is currently advocating for such a demonstration.[30]

Actual implementation of a full-scale Medicare-in-Mexico program would likely result in important behavioral responses, such as increased immigration. A Medicare demonstration, however, would not able to capture such behavioral effects.

Discussion

Ultimately, extending Medicare to eligible beneficiaries in Mexico could be a mutually beneficial endeavor that could strengthen the Mexican economy, reduce Medicare costs, and improve the lot of Medicare-eligible retirees in Mexico. It could also be a mechanism for building stronger ties and closer cooperation between the Mexican and American health bureaucracies. To overcome practical and political hurdles, a sophisticated model should be developed, similar to the one outlined in Box A, to build an evidence base for program costs and benefits under each of the policy options and to guide the pursuit of one or more policy options through a Medicare demonstration. Such a demonstration would likely take on a “life of its own” if the expected benefits predicted by the model were realized, making policy change for extending Medicare to Mexico not only feasible but also desirable across all relevant parties.

This article results from the RAND Corporation's Investment in People and Ideas program. Support for this program is provided, in part, by the generosity of RAND's donors and by the fees earned on client-funded research.

Acknowledgments

We would like to thank our colleagues Barbara Wynn, C. Richard Neu, and C. Ross Anthony for their insights and constructive comments on an earlier version of this document. We are also grateful to Jennifer Gelman for her help in our information gathering efforts and to Mary Vaiana for her editorial assistance.

Notes

1. This rule has a few, specific exceptions, including health emergencies arising in the United States but the closest hospital being foreign; health emergencies arising in Canada while traveling directly between Alaska and another state; and any medical needs if you live in the United States and a foreign hospital is closer to your home than the nearest U.S. hospital that can treat your medical condition (Centers for Medicare and Medicaid Services, Medicare and You 2010, 2010, p. 38. As of July 7, 2010: http://www.medicare.gov/Publications/Pubs/pdf/10050.pdf).

2. Centers for Medicare and Medicaid Services, Choosing a Medigap Policy: A Guide to Health Insurance for People with Medicare, 2010. As of July 7, 2010: http://www.medicare.gov/publications/pubs/pdf/02110.pdf

3. “TRICARE for Life,” Military.com, 2010. As of July 7, 2010: http://www.military.com/benefits/tricare/tricare-for-life/tricare-for-life-and-dual-eligibility#6

4. TRICARE website, no date. As of July 7, 2010: http://tricare.mil/mybenefit/

5. U.S. Office of Personnel Management, The 2010 Guide to Federal Benefits for Federal Retirees and Their Survivors, November 2009, p. 4. As of July 7, 2010: http://www.opm.gov/insure/health/planinfo/2010/guides/70-09.pdf

6. U.S. Government Accountability Office, 2010 Census: Counting Americans Overseas as Part of the Decennial Census Would Not Be Cost Effective, Washington D.C., GAO-04-898, August 2004. As of July 7, 2010: http://www.gao.gov/new.items/d04898.pdf

7. El Consejo Nacional de Población (CONAPO) Migration Series, “Población inmigrante residente en México,” no date. As of July 7, 2010: http://www.conapo.gob.mx/es/CONAPO/Poblacion_inmigrante_residente_en_Mexico. CONAPO Migration Series, Cuadro II.1.7, 2009. As of July 7, 2010: http://www.conapo.gob.mx/MigrInternacional/Series/02_01_07.xls

8. Social Security Administration, “Table 5.J11: Number and Total Monthly Benefits for Beneficiaries in Foreign Countries,” December 2008. As of July 7, 2010: http://www.ssa.gov/policy/docs/statcomps/supplement/2009/5j.html#table5.j11

9. Warner, David C., Medicare in Mexico: Innovating for Fairness and Cost Savings, Lyndon B. Johnson School of Public Affairs Policy Research Project Report Number 156, 2007.

10. Migration Policy Institute, America's Emigrants: U.S. Retirement Migration to Mexico and Panama, 2006, p. 24. As of July 7, 2010: http://www.migrationpolicy.org/pubs/americas_emigrants.pdf

11. Social Security Administration, 2008.

12. CONAPO Migration Series, “Población inmigrante residente en México,” no date. As of July 7, 2010: http://www.conapo.gob.mx/index.php?option=com_content&view=article&id=322&Itemid=251. CONAPO Migration Series, Cuadro II.1.7, 2009. As of July 7, 2010: http://www.conapo.gob.mx/MigrInternacional/Series/02_01_07.xls

13. CONAPO Migration Series, Cuadro II.1.6 and Cuadro II.1.7, 2009. As of July 7, 2010: http://www.conapo.gob.mx/MigrInternacional/Series/02_01_06.xls and http://www.conapo.gob.mx/MigrInternacional/Series/02_01_07.xls

14. Migration Policy Institute, 2006.

15. U.S. Census Bureau, “U.S. Population Projections,” 2008. As of July 7, 2010: http://www.census.gov/population/projections/

16. Croucher, Sheila, “Migrants of Privilege: The Political Transnationalism of Americans in Mexico,” Identities, Vol. 16, No. 4, July 2009, pp. 463–491.

17. Simcox, David, Another 50 Years of Mass Mexican Immigration, Center for Immigration Studies, March 2002. As of July 7, 2010: http://www.cis.org/MexicanGovernmentReport-50YearsMassImmigration

18. Aguila, Emma, and Julie Zissimopoulos, “Labor Market and Immigration Behavior of Middle-Aged and Elderly Mexicans,” Santa Monica, Calif.: RAND Corporation, WR-726, 2010. As of July 7, 2010: http://www.rand.org/pubs/working_papers/WR726/

19. Organisation for Economic Co-Operation and Development (OECD), OECD Reviews of Health Systems: Mexico, 2005. As of July 7, 2010: http://www.oecd.org/document/46/0,3343,en_2649_37407_35134254_1_1_1_1,00.html

20. OECD, 2005.

21. OECD, 2005.

22. We understand that foreign residents are able to buy into IMSS, but we have not been able to establish for certain whether Mexican citizens, returning to Mexico after working abroad, also have the option of buying in. We also understand that the cost of buying in varies from state to state and can run anywhere between $200 and $500 USD per year.

23. United States Citizenship and Immigration Services, Welcome to the United States: A Guide for New Immigrants, September 2007. As of July 7, 2010: http://www.uscis.gov/sites/default/files/files/nativedocuments/M-617.pdf

24. Croucher, 2009.

25. Social Security and Medicare Board of Trustees, Status of the Social Security and Medicare Programs: A Summary of the 2009 Annual Reports, 2009. As of July 7, 2010: http://www.ssa.gov/OACT/TRSUM/index.html

26. McGlynn, Elizabeth, Steven M. Asch, John Adams, Joan Keesey, Jennifer Hicks, Alison DeCristofaro, and Eve A. Kerr, “The Quality of Health Care Delivered to Adults in the United States,” New England Journal of Medicine, Vol. 348, No. 26, June 26, 2003, pp. 2635–2645.

27. Hawley, Chris, “Mexico's Health Care Lures Americans,” USA Today, September 1, 2009. As of July 7, 2010: http://www.usatoday.com/news/world/2009-08-31-mexico-health-care_N.htm

28. Migration Policy Institute, 2006.

29. Migration Policy Institute, 2006.

30. Association of American Residents Overseas, “Support Sought for Demonstration Project for Medicare Coverage in Mexico,” no date. As of July 7, 2010: http://aaro.org/index.php?option=com_content&view=article&id=144:aaro-endorses-proposal-for-medicare-demonstration-project-in-mexico&catid=23:medicare&Itemid=27