Problem with feasibility constraint

[makota]

Hey,

I have this model in which public sector collects taxes (T). Every period it gives f*T as transfers and and "saves" the rest (1-f)*T where f>0. The amount saved is given as transfers next period (with interest). Like this:

x = tax*w*l;
transfers = (1-f)*x + f*(1+r)*x(-1);
k = s + f*x;

Now here's the problem. I want to set a deterministic shock such that f=0. But what happens here, is that a part of capital kind of "disappears" so that the feasibility constraint ( y = k - (1-delta)*k(-1) + c)
does not apply in period 2 (if the shock is introduced in period 1). The feasibility constraint is wrong exactly by f*x. I was thinking about introducing a separate fund that is alive for 1 period only to fix the problem. But it doesn't work, as the agents also take all this into account. Anyone have any ideas how to go around this problem?

Cheers,
M

[jpfeifer]

How do you set a deterministic shock such that f=0?

[makota]

That's exactly my question. I don't know.

The usual way of...

shocks
var f;
periods 1:10;
values 0;
end;

..is no good. I'm guessing I would have to tweak the model somehow, but I'm stuck.

[jpfeifer]

Your timing seems to be wrong. Shouldn't it e.g. be f(-1)*(1+r)*x(-1) as the savings share should be the one from last period.

[makota]

Excellent. It works now! Thank you so much.