# Discussion Paper

## Abstract

All economists say that they want to take their model to the data. But with incomplete and highly imperfect data, doing so is difficult and requires carefully matching the assumptions of the model with the statistical properties of the data. The cointegrated VAR (CVAR) offers a way of doing so. In this paper we outline a method for translating the assumptions underlying a DSGE model into a set of testable assumptions on a cointegrated VAR model and illustrate the ideas with the RBC model in Ireland (2004). Accounting for unit roots (near unit roots) in the model is shown to provide a powerful robustification of the statistical and economic inference about persistent and less persistent movements in the data. We propose that all basic assumptions underlying the theory model should be formulated as a set of testable hypotheses on the long-run structure of a CVAR model, a so called ‘theory consistent hypothetical scenario’. The advantage of such a scenario is that if forces us to formulate all testable implications of the basic hypotheses underlying a theory model. We demonstrate that most assumptions underlying the DSGE model and, hence, the RBC model are rejected when properly tested. Leaving the RBC model aside, we then report a structured CVAR analysis that summarizes the main features of the data in terms of long-run relations and common stochastic trends. We argue that structuring the data in this way offers a number of ‘sophisticated’ stylized facts that a theory model has to replicate in order to claim empirical relevance.

## JEL Classification

## Cite As

## Assessment

# Comments and Questions

see attached file

see attached file

As many physicist involved by mistake in economic research I do not trust economic models exactly for the reasons discussed by Juselius and Franchi. There is no empirical validation. In the realm of mathematics, any non-contradictory set of assumptions potentially leads to a sound result, even if it does not ...[more]

... correspond to any observable processes or phenomena. In real world, and economy is a part of it, abstract mathematical consideration has to be justified by some observations - better direct ones, but indirect also work in many cases, especially at the edge of current knowledge. Economics, as a science, practically rejects such an approach from the very beginning, as one can learn reading university textbooks and modern economic and econometric papers. Such an approach can survive only in the absence of accurate and comprehensive concept justified and validated empirically, i.e. in the absence of any concurrence. Juselius and Franchi re-open this door (I guess in line with Frisch’s original ideas on econometrics) for establishing a new standard of economic and econometric modelling. This effort deserves strong support from both sides – economists and physicists. The next natural step would be a consensus procedure for rejection of those economic models which contradict observations, as in physics

The results obtained by Juselius and Franchi have actually uncovered another fundamental problem of economics as a science. There is a not well perceived contradiction between the current (invalidated) status of economic theory and relatively successful economic evolution in developed countries. If (macro)economic theories are not correct in many aspects why their application should be successful? For example, many if not all conventional theories say that investment is the driving force of economic growth as channeled through shocks to technology and total factor productivity. So, people must invest a certain portion of total output for an economy to grow. The paper puts the force under strong doubt – these variables evolve as a response to the shocks to demand for the consumption and for labor. Therefore the role of investment is not so clear now – one can originate consumption not only from new technologies.

Then, why are there no signs that investment suppresses observed economic growth? Imagine (what a nightmare for a physicist) that buildings, planes, bridges, and so on are constructed using not the experimentally tested and retested formulas but a theory like RBC or DSGE. It would be very difficult to find some economists, even a huge theoretical literature would be devoted to the safety as evaluated according to the RBC-like standards, using such empirically unreliable things. So, why don't economies suffer?

The answer is simple - actual economy is driven by forces beyond reach of economic and monetary authorities. Its evolution is not affected by weak actions advised by conventional economic theories. (Strong actions such as socialistic distribution and other means of oppression, of course, always at work to suppress an economy). Therefore, one has to look for a driving force of macroeconomic variables only in quantitative (not qualitative such as education, language, traditions in food and drinks, propensity to save, etc.) properties of population itself. It is obvious that no population means no economy.

In our papers we present a set of models describing such macroeconomic variables as real economic growth and inflation in developed countries using very parsimonious (one driving variable) models. These models are tested for cointegration in the VAR representation and meet the criteria formulated by Juselius and Franchi. We also consider the models as empirically justified since they describe actual measurements. They also provide an opportunity to predict evolution of inflation and real GDP at various time horizons. This is an appropriate basis for future validation.

There is no need to repeat full content of the papers here, so we just provide hyperlinks.

http://ssrn.com/abstract=969975

http://ssrn.com/abstract=960047

http://ssrn.com/abstract=956014