Reduced form

Exogenous variables are variables which are not determined by the system. If we assume that demand is influenced not only by price, but also by an exogenous variable, Z, we can consider the structural supply and demand model

supply:    ${\displaystyle Q=a_{S}+b_{S}P+u_{S},}$

demand:   ${\displaystyle Q=a_{D}+b_{D}P+cZ+u_{D},}$

where the terms ${\displaystyle u_{i}}$ are random errors (deviations of the quantities supplied and demanded from those implied by the rest of each equation). By solving for the unknowns (endogenous variables) P and Q, this structural model can be rewritten in the reduced form:

${\displaystyle Q=\pi _{10}+\pi _{11}Z+e_{Q},}$

${\displaystyle P=\pi _{20}+\pi _{21}Z+e_{P},}$

where the parameters ${\displaystyle \pi _{ij}}$ depend on the parameters ${\displaystyle a_{i},b_{i},c}$ of the structural model, and where the reduced form errors ${\displaystyle e_{i}}$ each depend on the structural parameters and on both structural errors. Note that both endogenous variables depend on the exogenous variable Z.

If the reduced form model is estimated using empirical data, obtaining estimated values for the coefficients ${\displaystyle \pi _{ij},}$ some of the structural parameters can be recovered: By combining the two reduced form equations to eliminate Z, the structural coefficients of the supply side model (${\displaystyle a_{S}}$ and ${\displaystyle b_{S}}$) can be derived:

${\displaystyle a_{S}=(\pi _{10}\pi _{21}-\pi _{11}\pi _{20})/\pi _{21},}$

${\displaystyle b_{S}=\pi _{11}/\pi _{21}.}$

Note however, that this still does not allow us to identify the structural parameters of the demand equation. For that, we would need an exogenous variable which is included in the supply equation of the structural model, but not in the demand equation.

Let y be a column vector of M endogenous variables. In the case above with Q and P, we had M = 2. Let z be a column vector of K exogenous variables; in the case above z consisted only of Z. The structural linear model is

${\displaystyle Ay=Bz+v,}$

where ${\displaystyle v}$ is a vector of structural shocks, and A and B are matrices; A is a square M  × M matrix, while B is M × K. The reduced form of the system is:

${\displaystyle y=A^{-1}Bz+A^{-1}v=\Pi z+w,}$

with vector ${\displaystyle w}$ of reduced form errors that each depends on all structural errors, where the matrix A must be nonsingular for the reduced form to exist and be unique. Again, each endogenous variable depends on potentially each exogenous variable.

Without restrictions on the A and B, the coefficients of A and B cannot be identified from data on y and z: each row of the structural model is just a linear relation between y and z with unknown coefficients. (This is again the parameter identification problem.) The M reduced form equations (the rows of the matrix equation y = Π z above) can be identified from the data because each of them contains only one endogenous variable.

• Simultaneous equations model#Structural and reduced form
• System of linear equations
• Simultaneous equations

• Dougherty, Christopher (2011). "Simultaneous Equations Estimation". Introduction to Econometrics (Fourth ed.). New York: Oxford University Press. pp. 331–353. ISBN 978-0-19-956708-9.
• Goldberger, Arthur S. (1964). "Reduced-Form Estimation and Indirect Least Squares". Econometric Theory. New York: Wiley. pp. 318–329. ISBN 0-471-31101-4.
• Klein, Lawrence R. (1974). "Regression Systems of Linear Simultaneous Equations". A Textbook of Econometrics (Second ed.). Englewood Cliffs: Prentice-Hall. pp. 131–196. ISBN 0-13-912832-8.
• Kmenta, Jan (1986). Elements of Econometrics (Second ed.). New York: Macmillan. pp. 651–660. ISBN 0-02-365070-2.
• Wooldridge, Jeffrey M. (2002). Econometric Analysis of Cross Section and Panel Data. Cambridge: MIT Press. pp. 211–215. ISBN 0-262-23219-7.

• Thoma, Mark (February 18, 2011). "Econometrics lecture (topic: structural and reduced form equations)". Economics 421/521. University of Oregon – via YouTube.