In responding to this question, one can use either an economic or a cultural perspective.

From an economic perspective, I would support a national exchange. There would be enormous economies of scale in information systems alone. National exchanges also would have more bargaining clout with insurers seeking to be listed on the exchanges. Finally, I do believe that there is virtue in uniformity in a geographic space whose inhabitants think of that space as a “nation,” and not just some territory shared by separate, diverse groups.

As a former longtime Board member of the Teachers Insurance and Annuity Association (TIAA), I recall the difficulties of getting even small changes dictated by changing market conditions through 50 separate state insurance regulators. The legal bill alone was staggering, as TIAA literally had to hire legal teams in every state to get any proposed change past state regulators.

Now, on the cultural plane, I have found myself bombarded for decades now with the mantra that, in America “one size does not fit all.” One hears it recited any time someone opposes a new federal law or institution.

If the mantra were valid, however, then explain to me general, cross-country uniformity of the United States. One Holiday Inn looks just like any other, and ditto for the Marriott hotels and McDonalds and Burger Kings, etc. The music on our radio stations is pretty uniform too – the same list of 30 top ditties player over and over again by disc jockeys with pretty much the same lingo -- even accent.

There are huge differences in the food, life styles and dialect and habits across, say, France and Germany. A German speaking the Hamburg dialect could not possibly understand someone speaking the dialect spoken in the Rhineland (around Cologne) or in Bavaria. Houses in bavaria look totally different from houses in Lower Saxony. By contrast, there are few cross-country differen ces U.S. life styles. On our first trip by car across the U.S., my wife and I were stunned by the almost dreary uniformity of our towns and cities: the same gas stations at major intersections, the same strip malls, the same fast food outlets, the same chruches, the same everything.

In fact, I believe the United States is the poster boy for the economic and administrative theory that one size fits all. It has been the very foundation of our stunning economic growth and of one of our greatest invention: franchising.

So, next time you hear that in America “one size does not fit all,” think of what all-American bullshine that really is!

For the sake of the nation, if we are to have health insurance exchanges, which we need, then I hope it will be a national exchange with state franchises, so to speak.

While there are some benefits to state exchanges, a nationally operated exchange, as called for in the House health reform bill, would provide greater regulatory leverage over insurance carriers--as well as greater control over premiums and costs--than individual state exchanges could provide.

As outlined in a recent Fund report on health exchanges by Timothy Jost, many states’ limited resources could limit their effectiveness in regulating state markets dominated by large insurers. And it’s important to note that states would still have a regulatory role to play under federal exchange option. State insurance regulators would work with the federal government to oversee and enforce requirements for participating plans, as well as those that do not sell policies through the exchange. The House bill also allows states to apply for a waiver to run their own exchanges under the same rules and requirements established for the national exchange.

Another advantage of a national exchange is that it would likely have lower overall administrative costs than 50 different exchanges, which would duplicate resources and administrative processes. As Jost and colleagues point out, some states may end up with very low enrollment in the exchanges, particularly if the individual and small-group markets are allowed to continue operation outside the exchange. Experts consider 100,000 to be the minimum size for a stable risk pool; certain states with small populations will be unable to reach this size. In addition, Alain Enthoven and colleagues, in a Committee for Economic Development paper, argue that an exchange should have at least 25 percent of the private insurance market to reduce adverse selection and attract carriers to participate in the pool.

The health reform legislation will be a federal law aimed at a problem that is national in scope--and a substantial amount of federal money in the form of subsidies will flow through the exchanges. The federal government, which has significant experience running insurance exchanges--including the Federal Employees Health Benefits Plan, Medicare Advantage, and Medicare Part D—should manage the country’s insurance exchange.

While basic principles of federalism don’t provide an unambiguous answer to whether insurance exchanges should be run by the states or the federal government, those principles can help us consider the relative merits of the two approaches.

In order to be effective, exchanges will need access to specialized expertise – actuarial, market research, and the like. Generally, the federal government is more able to obtain such expertise and it is more efficient for the federal government to do so than it is to have each state seek it on its own.

A single national exchange will be more efficient for large, national insurers. Even my small organization has employees who live in five different jurisdictions; we would probably find it easier to interact with one national exchange than five separate ones. Concerns regarding risk selection against the exchange are probably easier to mitigate in a national pool.

A key feature of an effective exchange is market clout. A large national exchange will have more ability to demand higher performance and possibly lower rates from participating plans. Of course this is a theoretical benefit only; the experience with Medicare Advantage shows that the federal government is perfectly willing to not exercise its clout when it doesn’t want to.

But there are other features of exchanges that point strongly in favor of state administration. Exchanges must provide customer services to enrolling individuals and firms. A federal bureaucracy is unlikely to be as responsive or accountable as a state-organized one. There will be significant movement of enrollees between the exchange and Medicaid, and since the latter is a state-administered program, coordination will be easier if the exchange operates at the state level. The need to integrate exchange coverage and children on CHIP also argues for state-based exchanges.

Some of the most innovating purchasing efforts designed to reduce health care costs are occurring at the state level. These initiatives—which involve payment methods, data collection and reporting, and systemic approaches to quality improvement—are most effective when they have the participation of all payers. It’s much easier to imagine a state-based exchange participating than a single, national one. It is also much likelier that state-based exchanges will contract with innovative local or regional plans.

There are important matters of fiscal federalism at stake as well. Since individual subsidies will be based on family income, the federal government will be responsible for the entire marginal cost associated with an exchange that fails to negotiate good rates. It is hard to justify full federal risk for the cost of state behavior.

It is worth noting that the federalism considerations of administration of the insurance exchange and the regulation of insurance have some overlap, but they are not identical. I’ve focused here only on the exchange.

As with most federalism debates, proponents on either side tend to set aside principles and base their view of which level of government should be responsible on their view of which level will do the “best” job—where best is in the eye of the beholder. Given the many parts of the health reform terrain that are unknown, that is a tough prediction to make.

Here is what can be said with some confidence.

1. The performance of state-based exchanges will vary substantially. The performance of a single national exchange will almost certainly fall somewhere between the top and bottom state on almost every dimension: consumer protection, effective rate negotiation, quality reporting, customer service, etc. Even with state variation, the constellation of changes anticipated in health reform will significantly narrow the interstate differences that exist today.

2. Without a substantial commitment of financial resources, exchanges will fail no matter what level of government carries out the task. Separate state exchanges will certainly cost more at the outset. And states will need substantial technical assistance to get their exchanges up and running. Given current state fiscal conditions, establishment of state-based exchanges will require federal funding.

3. The choice between state and national involves a series of tradeoffs. There is no “right” answer; there are only choices that emphasize one set of potential benefits over others.

Health insurance exchanges are one of the key components of the House and Senate health care reform proposals. They have the potential to make the insurance market much more competitive by putting in front of consumers a choice of plans from competing carriers that have been standardized by actuarial value (one offering per carrier for each of the actuarial values). This “head-to-head” competition is likely to lead to lower margins in this market. The exchanges, if done well, would also reduce the costs of distributing health insurance to individuals and small businesses. Given their critical role in the reform, it is important that effective entities be created.

State and federal approaches to exchanges each have their inherent advantages. State approaches would provide more innovation, since states could pursue distinct approaches and learn from each other. But state approaches can also lead to some very weak exchanges in some states. It is one thing for states like Massachusetts that have demonstrated their commitment to this approach to be relied on to perform this role. It is another to depend on states that do not support the concept, and that have myriad competing demands for resources and policy attention, to take on such a responsibility. Exchanges are going to have to be inventive in dealing with issues like minimizing risk selection and creating risk adjustment mechanisms. Additionally, creating and running 50 exchanges will cost a lot more than creating a federal exchange.

Turning to a federal exchange, the federal government does bring some meaningful experience to the task. The Office of Personnel Management has been doing this for decades for plans open to federal employees. It does not seem to have much trouble showing employees the local plans operating in their area as well as national plans available to all. CMS has experience managing Medicare Advantage as well as Part D Prescription Drug plans and is leading edge in performing risk adjustment. Certainly a federal approach would not preclude allowing states, such as Massachusetts, applying to run their own exchanges within the framework of a federal exchange.

Some have argued that states would do a better job at regulating insurers. I am not sure that is the case. Note that the federal government only has jurisdiction over benefits provided by large self-insured employers. Regulating fully-insured products might be a very distinct story. Consumer advocates have been critical of the job that a number of states have done in this area. Nevertheless, having a federal exchange does not preclude continuing with state regulation. Recall that exchanges are performing a function that is essentially a new one—structuring a marketplace where individuals and small employers can buy coverage.