Gerhard Tintner, A Simple Aggregate Model for Austria

Gerhard Tintner’s conference contribution, published in 1976 and represented here by a 2012 digital facsimile, develops a compact econometric model of the Austrian economy for policy analysis. Its central move is to connect aggregate consumption and production with national-accounting relations, then calculate how prices, income, output, employment, and consumption respond jointly to changes in externally specified variables. The contribution proceeds from definitions and equations through estimation methods to simultaneous elasticities and their interpretation, concluding with Austrian annual data for 1953–1972 and references.

The model is a highly aggregated Keynesian model.

This opening identifies both the theoretical orientation and the deliberate compression of the economy. Five variables are determined within the system: nominal private consumption, nominal gross national product, the price level, real gross national product, and employment. Population, public consumption, investment, inventory changes, exports, imports, capital stock, and nominal annual earnings enter from outside it. Tintner nevertheless acknowledges that these supposedly independent quantities are themselves dependent “to a certain degree.” Exogeneity is thus a working boundary of the model, not a claim that wages, investment, or trade are economically unconditioned.

The demand side begins with a linear relation between real consumption per head and real income per head. Nominal GNP equals private consumption plus public consumption, gross investment, inventory changes, and exports, less imports. Tintner groups these non-private-consumption expenditures into a single quantity, Q, and uses reduced-form regressions to recover the consumption function’s structural parameters. The estimated income coefficient is 0.533, with an intercept of 1471. Consumption therefore responds to income within a simultaneous system rather than simply being fitted as an isolated relationship.

The attempt to make the consumption function dynamic by introducing a lagged variable or a linear trend failed.

This brief admission establishes an important limit: the reported model remains static. Tintner offers no detailed account of those unsuccessful specifications, and the subsequent elasticities describe comparative responses rather than adjustment paths through time. Real output is obtained by deflating nominal GNP. A labor-demand condition then links the production side to wages by assuming that labor’s marginal product approximately equals the real wage.

Production takes the Cobb–Douglas form, with employment and real capital stock as inputs. Tintner avoids estimating both input coefficients through ordinary multiple regression because capital and labor data are highly collinear. Instead, following Klein, he uses the wage–marginal-product relation to infer labor’s output elasticity from labor’s income share. Constant returns to scale makes capital’s elasticity the complement of labor’s. The resulting coefficients are 0.471 for labor and 0.529 for capital. This is a consequential conceptual move: an economic restriction supplies information that the observed input series cannot reliably distinguish through regression alone.

The model’s purpose becomes explicit when Tintner turns from estimation to policy.

To apply this model for purposes of economic policy we regard the endogenous variables as goals, the exogenous variables, which can be controlled by the policy maker, as instruments.

The qualification concerning controllability matters: an externally specified variable is not automatically an available policy lever. Nevertheless, the goals-and-instruments framework organizes the remainder of the contribution. Because the nonlinear system does not permit the explicit solution in levels that Tintner seeks, he differentiates its equations and expresses changes relative to mean values. This yields a recursive calculation of simultaneous elasticities: the percentage responses of all five endogenous variables to a one-percent change in an external variable.

The elasticity table gives the model its practical content. A one-percent increase in public consumption raises prices by approximately 0.16 percent, nominal GNP by 0.31 percent, real GNP by 0.15 percent, and employment by 0.30 percent. Investment and exports similarly expand nominal and real activity, whereas imports have negative effects within the model’s expenditure structure. These results distinguish nominal expansion from additional real production: increased demand is distributed between prices and output rather than translated wholly into either.

Wages and capital reveal different interactions. A one-percent rise in nominal annual earnings increases prices by about 0.54 percent while reducing real output by 0.35 percent and employment by 0.72 percent. A one-percent increase in capital stock lowers prices by 0.57 percent and raises real output by 0.46 percent, yet slightly reduces employment, nominal GNP, and nominal consumption. These are conditional implications of the whole system, not independent causal estimates or unrestricted policy prescriptions.

The elasticities listed in Table 2 hold approximately in the range of a 5 % positive or negative change in the exogenous variables.

Tintner further conditions interpretation on the model’s approximate validity and on other external variables remaining unchanged. Its relevance lies in making aggregate policy trade-offs calculable with a small number of equations and transparent restrictions. Its limits are equally integral: static consumption behavior, constant returns, the wage-productivity assumption, and externally fixed magnitudes sustain the results. The contribution offers a locally applicable framework for tracing interconnected responses, rather than evidence that the reported elasticities remain stable across large changes or different economic circumstances.