Most of you will know of the “Phillips machine,” better called, as Mary Morgan suggests, the “Newlyn-Phillips machine,” a hydraulic model of a national economy.  It has a good literature, and while no complete mechanical description has been published, there are a couple of surviving machines one can examine.

But there was another machine.  In 1952, Phillips designed and built a mechanical device to link two national-economy machines, at either fixed or floating exchange rates, producing a two-country international economy model.  He seems to have done this as a one-off for James Meade, who used it in his LSE seminar from 1952 to 1957.

Linking two machines at a *fixed* exchange rate is fairly simple.  This is what was done, on Meade’s instructions, for a 1991 demonstration at LSE.  There's good documentation.  Linking at a *floating* rate is harder, and it’s not obvious how he did it.  Apart from three 1952 photographs, I have found no documentation of the Phillips foreign exchange machine in the LSE archives.  The physical device seems lost, as are Phillips’ drawings.

Any suggestions?  One possibility might be to find surviving students from Meade’s 1953-57 seminar with good mechanical memories, and I’d be grateful for any leads there.  Aside from the purely mechanical part, this connects to Meade’s policy interest in flexible rates during the 1950s.