This paper takes a parametric approach to demand analysis and tests the weak separability assumptions that are often implicitly made in representative agent models of modern macroeconomics. The approach allows estimation and testing in a systems-of-equations context, using the minflex Laurent flexible functional form for the underlying utility function and relaxing the assumption of fixed consumer preferences by assuming Markov regime switching. We generate inference consistent with both theoretical and econometric regularity. We strongly reject weak separability of consumption and leisure from real money balances as well as weak separability of consumption from leisure and real money balances, meaning that the inclusion of a money in economic models would be of quantitative importance. We also investigate the substitutability/complementarity relationship among different categories of personal consumption expenditure (nondurables, durables, and services), leisure, and money. We find that the goods are net Morishima substitutes, but because of positive income effects they are gross complements. The implications for monetary policy are also briefly discussed.

For previously identified weakly separable blockings of goods and assets, we construct aggregates using four superlative index numbers, the Fisher, Sato-Vartia, Törnqvist, and Walsh, two non-superlative indexes, the Laspeyres and Paasche, and the atheoretical simple summation. We conduct several tests to examine how well each of these aggregates “fit” the data. These tests are how close the aggregates come to solving the revealed preference conditions for weak separability, how often each aggregate gets the direction of change correct, and how well the aggregates mimic the preference ranking from revealed preference tests. We find that, as the number of goods and assets being aggregated increases, the problems with simple summation manifest.

In aggregation theory, the admissibility condition for clustering components to be aggregated is blockwise weak separability, which also is the condition needed to separate out sectors of the economy. Although weak separability is thereby of central importance in aggregation and index number theory and in econometrics, prior attempts to produce statistical tests of weak separability have performed poorly in Monte Carlo studies. This paper introduces a new class of weak separability tests, which is seminonparametric. Such tests are both based on a necessary and sufficient condition and are fully stochastic, allowing to take into account measurement error. Simulations show that the tests perform well, even for large measurement errors.

Weak separability is a key admissibility property in the Divisia approach to monetary aggregation. We test groups of U.K. household sector monetary assets for weak separability using new data underlying the Bank of England's benchmark revision of its household sector Divisia index. Nonparametric tests are used to identify four monetary asset groupings, which are weakly separable over all or almost all of the post-ERM period (1992:4–2005:1). We construct Divisia monetary aggregates for these four groupings and investigate their information content in two applications. The main findings are that Divisia money has direct effects on aggregate demand and that the growth rates of the nominal Divisia monetary aggregates Granger cause nominal output growth, but not inflation.

The Generalized Axiom of Revealed Preference (GARP) can be violated because of random measurement errors in the observed quantity data. We study two tests proposed by Varian (1985) and de Peretti (2004), which test GARP within an explicit stochastic framework. Both tests compute adjusted quantity data that are compliant with GARP. We compare and contrast the two tests in theoretical terms and in an empirical application. The empirical application is based on testing a large group of monetary assets for the United States over multiple sample periods spanning 1960–1992. We found that both tests provided reasonable results and were largely consistent with each other.