Based on two Romanian household surveys, we analyse the structure of households' income by sources: main job, secondary job, and hidden activities. After conceptual clarification and explanation of the methodology we used, we estimate the size of informal economy, analyse the relationship between variables related to different types of income, and explore the dynamics of the informal economy. We find that the main participants in the informal economy are the poor people: the survival motive is dominant in the Romanian informal economy. We estimate that both in September 1996 and in July 2003 the income from the informal economy amounted to about 1/4 of the total household income (23.6% in 1996 and 22.7% in 2003, respectively). Also, we estimate the share of income from the informal economy in the cases of various categories of population (defined according to the dimension of the official declared income per person in the household). The extension of our analysis to the entire year using the household population structure by deciles suggests that the informal economy has increased, on average, by about 2-2.5% over the period 1995-2002. Indeed, beside the actual level of income, the households' involvement in informal activities is probably influenced by occupation, region, age, education, number of children and many other factors. However, certain conclusions could be outlined: a) People perceive taxation as the main cause of the underground economy. b) Separating the main motivations of operating in the informal sector in two groups, “subsistence” and “enterprise” respectively, the surveys suggest that the subsistence represented a relevant reason for the households' decision to operate in the informal economy, including its underground segment. c) Informal activities supplied a “safety valve” within the surviving strategies adopted by the poorest households. d) Participation in informal economy seems to be not simply correlated with poverty: in the informal economy are involved poor people (having probably a low educational level), as well as rich persons, but their motivations are quite different. The former are practically “forced” to operate in the informal economy (the “subsistence” criterion), but the latter are “invited” to participate in it (the “enterprise” criterion). In both cases, at least during the first stages of transition to a free market system in Romania, the environment was propitious due to legislative incoherence, feeble penalty system in the cases of fraudulent activities, and existence of some accompanying elements of proper informal activity, such as corruption, bureaucracy, etc. However, the household's behaviour related to the participation in informal economy is sometimes fundamentally different between the two extreme groups of population. This is why in this study we focused on a deeper investigation of the behavioural aspects of different groups of population related to the implication in the informal sector.

Using adequate composite indicators, indeed together with other specific models, to analyse high frequency time series and to obtain sort-term forecasts can improve information for business environment, in modern era characterised by an accelerate process of changing. In our study we tried to build a composite indicator based on some monthly time series and to use it in order to obtain short-term forecasts for economic activity at national level. This indicator could be useful taking into account that actually there is no synthetic indicator to describe short-run dynamics of economic activity. To verify hypotheses of the estimating model for composite index, we used in case of Romanian economy the quarterly time series for the elements of it and quarterly published GDP as a benchmark indicator.

The study concentrated on demonstrating how non-linear modelling can be useful to investigate the behavioural of dynamic economic systems. Using some adequate non-linear models could be a good way to find more refined solutions to actually unsolved problems or ambiguities in economics. Beginning with a short presentation of the simplest non-linear models, then we are demonstrating how the dynamics of complex systems, as the economic system is, could be explained on the base of some more advanced non-linear models and using specific techniques of simulation. We are considering the non-linear models only as an alternative to the stochastic linear models in economics. The conventional explanations of the behaviour of economic system contradict many times the empirical evidence. We are trying to demonstrate that small modifications in the standard linear form of some economic models make more complex and consequently more realistic the behaviour of system simulated on the base of the new non-linear models. Finally, few applications of non-linear models to the study of inflation-unemployment relationship, potentially useful for further empirical studies, are presented.