Bicycles. That was the solution proposed by the British politician Norman Tebbit to the high unemployment of the 1980s. There was always a job somewhere if you looked hard enough.
Except there isn't always. Not even in a country as big as Britain. Now, 30 years later, it is easier to get on an aeroplane, and the scope for job-seeking is that much greater.
Perhaps it should be regarded as part of the economic natural law that people from a small country like Ireland treat the rest of Europe, and the world beyond, as their labour market.
It may be unrealistic to expect that Ireland can provide careers across today's varied world of work. But if we wish to maximise the number of jobs at home, and the chances of emigrants returning, we must change the way we measure economic success, and the policies required to achieve it.
That would be the case even in better times than these. Right now, the economy is not generating even the employment that it theoretically could. That requires different labour policies than those appropriate to a growing economy.
Above all, it requires some fresh thinking.
Such thoughts were prompted by some of the sideline blogs and comments in the controversy over the ESRI working paper on the costs of going to work. A fairly common one is that people want to work, and would not be deterred even by a modest reduction in disposable income.
Recent experience seems to support that view. Unemployment fell to 4% during the boom. Less than 2% of the workforce could be said to be unemployable, or truly unwilling to work.
The social welfare argument as a work deterrent also seemed to take a knock. During the bubble years, the lowest rate of social welfare was increased faster than average earnings, without any apparent effect in reducing employment.
But it may also be the case that people's response to financial incentives and disincentives may be different when jobs are scarce.
We can say with a lot of confidence that jobs will remain scarce.
History suggests that Ireland has always struggled with "the want of employment". Economic theory suggests this is not surprising in a small economy, most of whose consumer goods must be imported.
One of the many depressing things about the crash is that, despite its severity, it has had virtually no effect on the content of public discourse.
Perhaps the worst example of the walking dead in current politics is the idea that growth and employment will come from more consumer spending. Jobs will come from making it easier and cheaper to employ people in services, not by waiting for sales to rise to the extent that it becomes necessary to hire.
That other zombie, the Croke Park 'debate', was shaking its hoary locks again last week. It is the emphasis on Croke Park as a protector of incomes which awakens fears of a return to more or less permanent high unemployment and lack of opportunities.
Government is the largest provider of domestic services and directly employs about one in six workers. We are going to need every job which can be paid for from the economy's private earnings. Cutting public employment in such circumstances amounts to self-mutilation.
The value of the agreement is that it could make the public sector more efficient. The customers badly need that, and so do those who will be looking for work.
There will be nothing in it for either if the efficiencies are used to continue to provide government workers with real incomes which grow in line - or slightly more than - the growth in the economy.
It's still a bit of a secret, but not everybody in the past thought emigration was really such a bad idea.