Housing Assets In The Bank of Italy’s Survey Of Household Income And Wealth

Household Wealth Distribution in Italy in the 1990s

“…Comparisons with external sources, such as the national accounts, show that the quality of income and expenditure estimates in the SHIW is comparable to that of similar surveys in other countries: for instance, underestimation of disposable income is valued at around 30 percent (Brandolini, 1999; see also Cannari and Violi, 1995). Data on wealth, on the other hand, are typically less reliable and their accuracy tends to vary across different assets, misreporting being lower for tangible assets than financial assets.…”

“…On the contrary, correcting for non and underreporting has a mixed impact but predominantly in the direction of reducing dispersion. This pattern is the net outcome of two different effects (Cannari and D'Alessio, 1993): (a) the adjustment for non-reporting tends to reduce inequality, because all wealthy households declare they hold bank deposits and, most of them, government bonds; (b) the correction for underreporting leads to an increase in inequality, as the phenomenon matters more for those financial assets, such as private securities, investment fund shares and corporate equities, held to a much greater extent by the wealthy. 17 The null hypothesis of equality of the two indices is tested by the asymptotically standard normal statistic…”

“…• A model proposed by and refined by Cannari and D'Alessio (1993) is applied to correct for nonreporting and underreporting of financial assets. The method is based on the outcome of a statistical matching of the SHIW data for 1987 with the micro-data from a survey carried out in the same year by the Banca Nazionale del Lavoro (BNLS) on a sample of its customers.…”

“…Indeed, after allowing for the different composition of the two samples, found that SHIW figures fell short of the corresponding BNLS aggregates by about a half, underreporting being higher for the households of the elderly, the less educated, and the self-employed. As adjustment for underreporting proposed by requires the availability of both the SHIW and the BNLS at the same time and no further BNLS has been carried out since 1987, we apply the updated and revised methodology developed by Cannari and D'Alessio (1993).…”

Italian Household Wealth: Background, Main Results, Outlook

“…Despite the considerable attention paid to strategies aimed at gaining the trust of respondent households, 6 the survey is affected by problems of selection bias, in other words by the lower levels of participation of wealthier households (D'Alessio and Faiella, 2002), and by under-reporting, i.e. statements regarding income or wealth that are not entirely truthful (Cannari and D'Alessio, 1990;1993;D'Aurizio et al 2006). Despite these problems, the surveys have proved to be an irreplaceable instrument of economic analysis, as testified to by the over 500 scientific works produced by more than 350 Italian and foreign authors listed in the bibliography of works based on the survey data (Biancotti and D'Alessio, 2007).…”