Solve for steady state

545 Views Asked by At

I am trying to solve for the steady state (in the context of a DSGE economic model) and one of the equations is reffering to capital accumulation. Particularly:

$$K_t=(1 - \Delta)K_{t-1} + \left( 1 - \frac{k_i}2 \left( \frac{I_t}{I_{t-1}} - 1 \right)^2 \right)I_t$$

With

  • $K_t ={}$ capital
  • $d = {}$ depreciation
  • $k_i = {}$ adjustments costs
  • $I_t = {}$ Investment

I am substituting $I_t = \Delta K_t$ (a general calibration advise) and try to solve in respect for $K$'s steady state. I am concluding $K=K$ which means that there is an infinite number of solutions. Is this statement correct? For example can I use $K=1$ and try to solve the rest system of equation for the steady state. Is this approach correct?

1

There are 1 best solutions below

2
On

You are approaching this incorrectly. In the steady state all variable should be constant. So you should impose that $K_t=K_{t+1}=K$ and $I_t=I_{t+1}=I$. If you plug this into your equation, you obtain $K=\delta I$. So the conclusion is that in the steady state, the investment just covers the depreciation of existing capital. But you still need to solve for $K$.

To solve for $K$ you have to consider the whole system of equilibrium conditions. This will typically be a non-linear system which will have to be solved numerically.