The hypothesis put forward in the abstract of this paper is that workers proportionately withdraw effort as their actual wage falls short of their fair wage.
This seems a sensible assumption to me but the hypothesis itself is not an easy one to test. Measuring the productivity of a workforce can be difficult and productivity levels can decrease for a number of different reasons. The main difficulty when testing this hypothesis would be finding a suitable environment in which tests could be undertaken. Safeguards would have to be put in place to measure the impact of any other factors on productivity as well as wage level.
With regard to Terry's question 'what does it mean when we say a hypothesis must be testable?' - An untested hypothesis is merely an opinion. If a hypothesis cannot stand up to scrutinisation then it is clearly flawed, although perhaps not entirely useless as it could well still add to a debate.
In the case of this particular paper there is bound to be debate over what a worker's fair wage actually is and how it should be determined. The socio-geographical factors as well as political arguements involved in such a debate would be very interesting indeed.
