The problem--well, one of many problems, actually--is that since employers are under no obligation to provide any level of insurance coverage, or indeed to provide any coverage at all, companies that are stingy with health care benefits will have a competitive advantage over companies that provide adequate insurance. Again, in theory, the market is supposed to fix this; employers who don't provide insurance will find it more difficult to find people willing to work for them, or they'll discover that the people who do work for them will demand higher wages in order to compensate for the lack of health benefits. But--surprise!--that doesn't seem to work very well in a lot of industries, particularly with so-called unskilled workers and particularly when the company is not unionized.
The result is that the Wal-Marts of the world can undersell their competitors, driving out companies that provide decent health insurance and forcing workers and their families to rely on public resources. In other words, the taxpayers end up paying for health care, but in a very inefficient way, for example when people with full-time jobs turn up at emergency rooms for non-emergency cases. Federal, state, and local governments spend $45 billion on Medicaid in New York State alone, and many of the people who use public health benefits and facilities (Medicaid and otherwise) are employed full-time. Which makes you wonder why we don't pay for health insurance with tax revenue in the first place and do it systematically rather than haphazardly, while also getting rid of one of employers' biggest expenses and biggest headaches (another medical benefit?).
If we don't pay for everyone's insurance and take the whole thing out of employers' P and L statements, there's another logical way to prevent cost competition from taking insurance away from entire industries, and New York City is poised to adopt that approach if the City Council can override Mayor Bloomberg's expected veto: require employers to provide a certain minimum level of insurance coverage.
According to this article, the Maryland legislature was also ready to try this solution, but Governor Ehrlich (who is rapidly becoming one of my least favorite governors even though I've never lived in Maryland) vetoed it.
This presentation explains the economic rationale for and the effect on employees of the Wal-Mart model, concluding that the success of the model on economic terms means that non-market solutions (e.g., regulations, enforcement of existing wage and labor laws) are needed.