Sunday, November 16, 2014

Why social democracy can work in a city, county, state, or country of any size

A common argument holds that social democracy cannot succeed in the United States because it is thirty times more populous than Sweden--population 10 million--the largest social democratic country (though Kerala is a social democratic state in India with 20 million residents). The reason why size is so important is never articulated--but it is nevertheless taken as obvious that social democracy is practical in a nation of 10 million, but impossible in a nation of 316 million.

There are several problems with this argument, however. From the outset, the United States has two extremely successful social democratic programs--Social Security and Medicare--so the burden of proof for this argument is enormous.

Indeed, this assumption contradicts the basic statistics behind social policy design. As I wrote previously, the major strength of social democracy is that the cost of providing services to each participant falls as the number of participants increases. Though there are diminishing returns, for social democratic-style social welfare programs, bigger is always better. The more people participating in a social democratic system, the more secure, efficient, and cost-effective that system will be. In theory, then, social democracy should function better in larger countries like the United States than in smaller ones like Denmark or Sweden.

Of course, statistics and actuarial tables occur in paper, ink, and computer models, not in the real world. Perhaps there are practical, more down-to-earth logistical or management difficulties that are incurred in administering social welfare programs when there are hundreds of millions of participants, rather than millions.

Indeed, intuition argues that large systems cannot be efficient. When a system gets larger and larger, more and more layers of bureaucracy and management are needed to successfully run that system. For an example, let's assume that every ten people need to be managed by a supervisor. For a program with ten employees, one supervisor is needed--that's one layer of management. But if there are 100 employees, then ten supervisors are needed--meaning those ten supervisors need a supervisor--and suddenly we have two layers of management. As more and more employees are needed to administer a program, more and more layers of management are needed, stifling efficiency with oppressive bureaucracy. At some point, the increasing efficiency of more enrollees is certainly overcome by the decreasing efficiency of needing to layer more and more levels of management.

But the intuition is wrong. The administrative costs of Social Security are less than 1% and those of Medicare are around 3%. Neither Social Security nor Medicare have layers upon layers of management. Medicare, for example, is administered by the Centers for Medicare and Medicaid Services (CMS). CMS is responsible for Medicare, Medicaid, HIPAA, parts of Obamacare, clinical laboratory standards, long term care facility standards, and the Children's Health Insurance Program (CHIP). All told, 100 million Americans are covered by a CMS health insurance program: Medicare, Medicaid, or CHIP. Despite enrolling 100 million people--nearly a third of the entire country, or ten times the entire population of Sweden--as well as having numerous other responsibilities, CMS only employs around 4000 people (far fewer than private sector insurance companies).

The Social Security Administration (SSA) covers nearly all 316 million Americans and includes programs as wide ranging as child welfare and disability benefits. SSA employs 60,000 people, which may seem like a lot--except that most of these people are responsible for running the 1,230 local field offices across the entire country (it's also worth noting that verifying disability claims is very labor-intensive). This is actually a very special feature of SSA--there are so many local field offices that any American can easily reach a field office in order to work with an SSA employee in person to ensure their concerns are resolved. SSA employees may help people with lost or stolen social security cards, retirement, disability, some Medicare eligibility issues, immigrants in need of social security numbers, SSI, etc. Nearly all of these things can be done by mail or over the Internet, but it's worth knowing that any concern that cannot be redressed by these means can be brought to the attention of a real live human being, face-to-face. Few national entities in the private sector offer this level of customer support.

But how can this be? Health insurance is very complicated. How can CMS cover 100 million people with just over 4000 employees? Clearly, social democratic programs have discovered a way to administer programs with minimal staff, somehow without layers upon layers of management.

The key is that social democratic programs are funded at the federal level but administered at the local level. The federal government sets rules and standards for service provision, but all service providers work for local entities--not federal ones. For example, there are no health care providers that actually work for Medicare or CMS. CMS funds all Medicare services, but all Medicare services are delivered by local hospitals and clinics. CMS only needs to verify that local providers don't commit billing fraud and comply with HIPAA patient privacy and other regulations. Beyond that, CMS has no need to manage the day-to-day affairs of health care providers. There's no need to have layer upon layer of management, because there's nothing really to manage. CMS has no need to manage day-to-day service provision. This marriage of federal funding and local administration is the key to the ultra efficiency of social democratic programs.

Indeed, this model of federal mandates and funding with local administration is the key to the success of social democratic programs in the social democracies. The social democracies may be much less populous than the United States, but they're still nations of millions of people. Their populations are too big for the federal government to practically manage service provision, so they, too, rely on the model of federal funding paired with local service administration. For example, in section 3 of this paper, Havnes & Mogstad detail the development of universal child care in Norway in the 1970's, noting that the federal law was specifically designed to be funded federally but implemented at a local level--which it was with great success.

For a broader, systems-based perspective, Elizabeth Bradley and Lauren Taylor recently wrote a book detailing their qualitative study of the social welfare systems in different countries. In this study, they examined how social welfare administration--for services as wide-ranging as health care, child care, ageing, education, long term care, and other services--affect health outcomes and the cost of health care. In an interview for this book, they explain that they found federal funding with local administration to be the rule for all types of social welfare services in Scandinavia:
While Scandinavia is often seen by the United States as a very socialized country, the truth is that they run on localism. All of this joint budgeting and planning happens at a local, county level...In Scandinavia, there's a sense of accountability for population health at the local level. I think that's part of the concern here. In Scandinavia, the local government really takes the lead and they're the arbiter of different interests. They make sure the process is moving forward, and that the planning and the budgeting is responsive to public needs.

Who's going to play that role in the United States? Someone needs to hold the reins. In Scandinavia, it's very clear [that local government] should lead.
I wrote here about some serious caveats in their understanding of these observations in a historical context. Nevertheless, it is clear that the social democracies rely on the federal funding, local administration model.

Again, this federal funding / local administration model has already been successfully scaled up to work in the Social Security and Medicare programs of the United States, with a population thirty times that of Sweden. The model works in small countries like Norway; the model works in large countries like the United States. Using this model for other social welfare services--like child care, health insurance, more comprehensive long term care insurance, parental leave, etc--there is no reason we could not have a comprehensive social democratic welfare state in the United States, despite the much larger population.

No comments:

Post a Comment