Description

This is a theoretical optimizing model of a typical household.
A detailed description can be found in Chapter 3 of Ray Fair's book.

Small Model of Type : NLP

Category : GAMS Model library

Main file : hhfair.gms

$title Household Optimization Model by Fair (HHFAIR,SEQ=69)

$onText
This is a theoretical optimizing model of a typical household.
A detailed description can be found in Chapter 3 of Ray Fair's book.


Fair, R, Specification, Estimation, and Analysis of Macroeconomic
Models. Harvard University Press, Cambridge, Mass, 1984.

Keywords: nonlinear programming, household optimization, macro economics
$offText

Set
   tl     'long time horizon'  / 0*3 /
   t(tl)  'optimizing horizon' / 1*3 /
   tt(t)  'terminal year'      /   3 /;

Scalar
   p      'price of goods'                              /    1.0      /
   r      'one period interest rate'                    /     .07     /
   tr     'transfer payment'                            /    0        /
   th     'total number of hours in the period'         / 1004.72366  /
   w      'wage rate'                                   /    1.0      /
   d      'income tax rate'                             /     .2      /
   a1     'distribution coefficient in ces function'    /     .5      /
   a2     'elasticity coefficient in ces function'      /    -.5      /
   am     'terminal assets target'                      / 1100        /
   lambda 'discount rate'                               /     .944    /
   lstar  'maximum labor available in each time period' /  400        /
   gamma1 'coefficient 1 in money holding function'     /     .255905 /
   gamma2 'coefficient 2 in money holding function'     /    1.0      /;

Parameter ufact(t) 'utility function factor';
ufact(t) = power(lambda,ord(t) - 1);
display ufact;

Variable
   a(tl) 'assets'
   c(t)  'consumption'
   l(t)  'labor supplied'
   m(tl) 'money holdings'
   n(t)  'time spent on money holdings'
   obj   'objective function value'
   s(tl) 'savings'
   tax(t)'net taxes paid'
   u(t)  'utility'
   y(t)  'income';

Equation
   objective    'objective function'
   utility(t)   'utility in each period'
   income(t)    'before tax income'
   taxes(t)     'net taxes'
   savings(t)   'savings'
   budget(tl)   'budget constraint'
   timemoney(t) 'time spent taking care of money holdings'
   terminal(t)  'terminal condition for assets'
   dom1(t)      'domain constraint on timemoney equation'
   dom2(t)      'domain constraint on utility equation';

objective..    obj    =e= prod(t, u(t)**ufact(t));

utility(t)..   u(t)   =e= (a1*c(t)**(-a2) + (1 - a1)*(th - l(t) - n(t))**(-a2))**(-1/a2)/100;

income(t)..    y(t)   =e= w*l(t) + r*a(t);

taxes(t)..     tax(t) =e= d*y(t) - tr;

savings(t)..   s(t)   =e= y(t) - tax(t) - p*c(t);

budget(tl-1).. s(tl)  =e= a(tl) - a(tl-1) + m(tl) - m(tl-1);

timemoney(t).. n(t)*(m(t) - gamma1*p*c(t)) =e= gamma2;

terminal(tt).. a(tt) + m(tt) =e=  am;

dom1(t)..      m(t) =g= 1.01*gamma1*p*c(t);

dom2(t)..      l(t) + n(t) =l= .9*th;

Model hh 'household optimization model' / all /;

l.up(t)   = lstar;
a.fx("0") = 1000;
m.fx("0") = 100;
c.lo(t)   = 100;
l.lo(t)   = 100;
u.lo(t)   = .01;

a.l(t) = 1000;
m.l(t) = 100;
l.l(t) = 400;
u.l(t) = 1;

solve hh maximizing obj using nlp;