RBC labor friction model: steady state problem

Hi everyone,

I’m novice in dynare, so please excuse me if this topic had been already treated.

I have to make a RBC model for a class project, i think i’ve got a problem with my exogeneous variable who can’t let me calculate my steady state.

Please if someone can check my code, i will be greatful.

My code:

//Baseline model in levels including capital
var c y k z w u l n r i v theta q f;

//y     output
//k     capital
//i     investment
//l     non particaption
//c     consumption
//r     interest rate
//n     employment
//u     unemplyment
//f     job finding rate 
//v     vacancies
//q     probability of filling a vacancy
//theta labor market tightness
//w     wage

varexo e;

parameters A beta s delta kappa alpha b rho phi eta mu nu csigmaepsilon khi;
A = 1.0;
beta = 0.9992;
s = 0.192;
delta = 0.0026;
kappa = 5.0;
alpha = 0.33;
b = 1.43;
rho = 0.9895;
phi = 0.90;
eta = 0.5;
mu = 0.053;
nu = 2.1;
csigmaepsilon = 0.0034;
khi = 0.0081;

model;

//RBC equations
c^(-1) = beta*c(+1)^(-1) * ( r(+1) + (1-delta) );                                //Euler for capital  
r = alpha * y * k(-1)^(-1);                                                      //r=marginal product
k = (1-delta) * k(-1) + i;                                                                 //LOM for capital                                             //Output                                                                    
y = c + i + kappa * v;   
y = A * z * n(-1)^(1-alpha) * k(-1)^(alpha);                                                                    //Accounting identity
z = rho*z(-1) + e;

//Labor-part equations
f = s * (theta)^(1-eta);
q = s * (theta)^(-eta);                     
theta = (v/u);                                // Labour market tightness
n=(1-khi) * n(-1) + (f * u);                  //
u = 1 - n(-1) - l;                               // Definition of uneployment
kappa/q=beta * c/(c(+1)) * ((1-alpha) * (y(+1)/n) - w(+1) + (1-khi) * (kappa / q(+1)));  // Job creation condition
w=(1-mu) * b * c + mu * ((1-alpha) * (y/n(-1)) + kappa * theta);                          // Aggregate wage
phi * l^(-1/nu)=  b + (f * beta * ((w(+1)/c(+1))-(phi * l(+1)^(-1/nu)) * ((1-khi)/f(+1)) * (phi * l(+1)^(-1/nu)-b)));
end;

initval;
c = 4.24;
y = 5.776;
k = 5.57;
w = 6.24;
u = 0.0352;
l = 0.36;
n = 0.6049;
r = 0.0034;
i = 0.0145; 
theta = 0.523;
q = 0.266;
f = 0.139;
v = 0.0184096;
z = 0;
end;

shocks;
var e = csigmaepsilon^2; 
end;

steady;
stoch_simul(order=1,irf=100);

The equation
z = rho*z(-1) + e;
implies that z has a steady state of 0. But this is inconsistent with
y = A * z * n(-1)^(1-alpha) * k(-1)^(alpha);
In that equation, you should have
y = A * exp(z) * n(-1)^(1-alpha) * k(-1)^(alpha);

Thanks but i had test with exp(z) and nothing change, octave said problem Blanchard khan…

The first step is to get the steady state right. Only then can you check the Blanchard-Kahn conditions. The latter are usually about timing problems