The types of people who travel by train rather than air shuttle are, on the whole, those who already know only too well about the decline of America's traditional industries. The acres of dereliction running alongside the tracks are a graphic illustration of it.
Yet the indications are that the US economy as a whole is in very good shape. America is now a low-inflation country: the average rate of consumer price inflation has been well below 4 per cent for a decade. Economic expansion has slowed but growth is expected to recover to near its trend rate.
The part that has lately been doing well enough to compensate for the decline of industries that no recovery can reach is the service sector. During the second quarter, when gross domestic product growth dipped to an annual rate of 1.3 per cent, service output was up 3.8 per cent, industrial output only 0.5 per cent.
The number of jobs in industry fell in the second and third quarters - by 18,000 in September alone. But the number employed in the service sector slowed only modestly, and private sector services added 169,000 jobs last month.
This is not a new trend. Service sector employment has grown faster than industrial employment for most of the past 10 years.
Service industries are thriving in places far from the east coast. For instance, California and the west coast have been boosted by the entertainment and communications industries, Georgia by the growth of financial services, Florida by catering to its booming population of wealthy pensioners.
The continuing switch from manufacturing to services is only one part of the explanation for America's strong economic performance. For there is no question that some areas of US manufacturing have been tempered in the forge of recession and international competition into world-class, ultra-competitive industries. Thanks to cost cutting and lay-offs, economy- wide productivity growth has been a spectacular 2 per cent a year since mid-1990, and in manufacturing about 3 per cent a year.
Higher growth of manufacturing output is going to be the key to a short- term pick-up in the economy, just as it was the explanation for the recent slowdown. Although it accounts for less than a fifth of the economy, the swings in manufacturing drive the economic cycle.
The latest statistics are not all that encouraging. Apart from the declines in manufacturing employment, the index of manufacturing activity produced by the National Association of Purchasing Managers has been below 50 - pointing to a fall in activity - in four out of the past five months. On all but one occasion in the last 26 years a run like this has heralded a recession. (The exception was 1985, a year of very expansionist monetary policy.)
However, no matter what happens in the short-run - and the bond market will react sharply if the economy diverges in either direction from its ''soft landing'' ideal of steady 2-2.5 per cent growth - America's long- run growth trend depends on services.
There are opposing forces acting on service industries, which make it extremely difficult to predict how strong their performance will be.
Services have traditionally displayed very sluggish growth in productivity - partly just because of difficulties in measuring it but partly because technological progress has not, so far, worked its magic on services. This could be about to change.
There is strong anecdotal evidence that computers are starting to boost productivity in a range of service industries. This is obvious in areas such as communications and entertainment but it also applies to distribution and retailing, for instance.
What's more, this recovery has seen a huge surge in investment in computers. Investment in information systems has grown at an annual rate of about 20 per cent since early 1993. The US now has 63 computers for every 100 workers, a ratio nearly four times as high as in Japan.
Unfortunately there are a lot of other services whose prospects are not nearly so dynamic. These include the bloated and inefficient health-care sector and the less bloated but even more inefficient public services, the obstructiveness of whose employees helps explain why so many Americans think their government is out to get them.
These sectors will face an unpleasant shake-up during the next few years. The outcome of the game of budget poker being played between the Clinton administration and Congress will be big cuts in Federal government spending.
A recent analysis by Darwin Beck, an economist at the broker Credit Suisse First Boston in New York, makes it clear that both sides agree that welfare and medical spending will bear the brunt of cuts during the next seven years. The wrangling in Washington is about precisely how much and how it is done.
Under the deficit reduction plans, the Federal government will devolve responsibility to State governments for some welfare programmes - Medicare (for the elderly) and Medicaid (for the poor). States are not permitted to run deficits on current spending.
This devolution will therefore enforce spending reductions, which will result in some combination of reduced services and improved efficiency.
There are bound to be job losses in the affected services and some economic weakness - which might be offset by bonuses such as lower long-term interest rates if the financial markets reward budget cuts. It is impossible to tell how severe the disruption to the economy might be, Mr Beck concludes.
Eventually, however, a shake-out might just boost productivity growth in the affected services. Like other service-dominated developed economies, the US certainly needs to find ways of raising growth in traditionally low-productivity sectors.
This is not simply a matter of raising the average growth rate. It is necessary in order to prevent greater inequality and to tackle the country's serious social divisions. There are already communities that have been catastrophically impoverished by de-industrialisation, alongside the successfully re-invigorated manufacturing companies.
The biggest economic policy challenge in the US is to ensure that developments in the far more important service sector do not similarly reinforce the gap between haves and have-nots - or travellers by air and by Amtrak.