Instead of relying on price stickiness, Devereux, Head and Lapham [1996] rely on
monopolistic competition in the intermediate good sector to obtain demand effects from
government spending. An increase in the latter causes entry in the intermediate good
sector and raises total factor productivity. With enough monopoly power in the intermediate
good sector, the real wage increases even though labor supply shifts out, by the usual
negative wealth effect. Private consumption increases; private investment also increases
because of the large increase in labor supply due to the increase in the real wage.
In the neoclassical model, usually the real interest rate increases in response to a
government spending shock. Intuitively, the marginal product of capital increases as
labor supply shifts out; and it increases more, the more persistent is the shock, because
the larger is the wealth effect and the accompanying outward shift in labor supply. The
real interest rate also increases in the neo-keynesian models of Linnemann and Schabert
[2003], Galí, López-Salido and Vallés [2003], and Devereux, Head and Lapham [1996].48
Note that all these results refer to the short interest rate; if the long interest rate is
extrapolated from the sequence of short interest rates, however, it would rise too. Thus,
in these models it is difficult to rationalize the small or negative response of the real
interest rate observed in S1.