It's accepted wisdom that rising medical costs will be a crushing burden for the next generation. The latest report from Medicare, for example, projects that its cost will rise from less than 3% of gross domestic product today to almost 14% in 75 years.
There's also little faith that technological advances in medicine will make a dent in these rising costs. Instead, new techniques seem to require enormous expenditures. And while science has made big strides in unraveling the human genome, the resulting new drugs are expensive to produce.
But before concluding that technology can't reduce costs, let's remember a bit of history. Throughout the PC boom of the 1980s and early '90s, the same doubts were raised about the payoff from info tech. People saw huge sums being spent on computers and software -- well over 3% of GDP -- with no apparent efficiency gain. Economists called this "the productivity paradox" and even proposed that computers were cutting efficiency, by giving people more opportunity to goof off at work. In 1987, Nobel prize winner Robert M. Solow famously said: "You can see the computer age everywhere but in the productivity statistics."
The payoff from info tech was there, however -- it just took longer than expected. Intel Corp. (INTC ) introduced the first microprocessor in 1971, but it took 25 years for real productivity gains to show up. Now they are obvious to everyone.
Today, biotech may well be where info tech was in the 1980s: a technology that is still expensive and hasn't matured yet. Genentech (DNA ) only brought the first recombinant DNA drug to market in 1982. If biotech follows the same path as info tech, then real gains in medical productivity -- measured by cost per patient -- should arrive toward the end of this decade.
While the nature of technological change is hard to predict, we can already see a glimmer of how these gains might happen. Improvements in biotech manufacturing, for example, are likely to bring down costs. Understanding better why only some cancer patients benefit from a treatment could make doctors more efficient in prescribing expensive drugs. And the continued shrinkage in the size and expense of devices such as implantable cardiac defibrillators could help limit hospital stays.
Current forecasts don't factor in the cost-reducing potential of such new technologies. Instead, most analysts assume that the pattern of recent years -- of health-care costs outracing GDP -- will continue indefinitely. That's why long-term projections show health care rising explosively as a share of GDP.
But in a dynamic industry such as health care, extrapolating current trends into the long-term future is a mistake. To see how ludicrous these assumptions are, let's apply the same methods to the future of info tech. In the past 10 years, business spending on technology, per person, has risen at 6.6% a year, compared with a 4% rise in per capita GDP. If we follow the lead of health-care analysts and extend this trend, we reach the absurd conclusion that, in 75 years, information technology will absorb almost 29% of GDP.
Moreover, most long-term forecasts use a projection method that rules out the possibility that the U.S. can grow its way out of the health-care-spending squeeze. The Medicare Trustees, like most other analysts, assume that over the long run, health-care costs per beneficiary will grow at a rate one percentage point faster than per-capita GDP. For example, if per-capita GDP rose at 3.5% a year, analysts assume medical costs would climb 4.5%. Or if GDP jumped an astronomical 5.5% a year, medical costs would surge 6.5%.
Under these assumptions, the economy is like a greyhound chasing a mechanical rabbit -- it can never catch up with medical costs. That's why most forecasts conclude that faster economic growth actually creates bigger Medicare deficits. In reality, however, there's no reason why more robust growth will automatically translate into even more medical spending.
In fact, the biggest dangers are a slowdown in growth and a lack of technological innovation. A long-term reduction in the rate of economic growth could lead to a smaller pie, making it harder to pay for medical care. And a lack of technological breakthroughs could leave today's labor-intensive, hospital-based system still in place -- an untenable long-term setup.
Nobody is saying that health-care costs are a minor issue. But the burden of paying for Medicare and medical care in general may turn out to be lighter than people expect.
By Michael J. Mandel