ICP 2005 Construction Prices: Are They Underestimated in Developing Countries?

Abstract

Construction prices are lower in developing countries in ICP 2005, which has raised these countries' capital/output ratio in Penn World Tables 7 and 8 and affected growth analyses. We estimate the Colombia/U.S. price ratio for office and apartment buildings in 2005 as a test of the validity of the ICP 2005 methodology for estimating these prices. Our estimate of the Colombia/U.S. price ratio is almost twice the ICP 2005 estimate. We confirm the validity of our results by estimating the cost of constructing office and apartment buildings using 2005 prices for construction materials, equipment, and labor in both countries. JEL Codes: E30, O47

A replication of ‘The long-run impact of foreign aid in 36 African countries: Insights from multivariate time series analysis’ (Oxford Bulletin of Economics and Statistics, 2014)

“…Building on a recent literature that uses human nighttime light emissions as a proxy for economic growth (most notably, Henderson et al, 2012) they provide estimates of the relative accuracy of different vintages of PWT as well as WDI. In contrast to Breton and García (2016), their findings suggest that the 2005 ICP round provides an improvement over earlier rounds of the survey. Nevertheless, their results corroborate the frequent finding that 'newer need not be better', in the sense that PWT 7.1 consistently outperforms the more recent PWT 8.0 and 8.1 series.…”

“…They show that this is mainly driven by changes in the methodology used to estimate prices in the construction sector: ICP 2005 bases its estimates on the prices of inputs, while earlier versions of the ICP estimated the prices of construction projects as such -a more demanding, but arguably more precise procedure. Breton and García (2016) come to conclude that earlier versions of the PWT (versions before PWT7) were superior in this respect. The most recent round of the ICP, based on prices collected in 2011, again differs substantially from the 2005 vintage.…”

“…One way of addressing this question is to examine the assumptions and methodologies underlying the datasets, and to assess their validity. For instance, Breton (2015) and Breton and García (2016), show that the main difference between ICP 1996 and ICP 2005, reflected in the Penn World Tables' PPP estimates, are the estimated prices of investment in developing countries. They show that this is mainly driven by changes in the methodology used to estimate prices in the construction sector: ICP 2005 bases its estimates on the prices of inputs, while earlier versions of the ICP estimated the prices of construction projects as such -a more demanding, but arguably more precise procedure.…”

“…Partly as a consequence of the low quality of the National Accounts data reported by national statistical offices, there can be large differences between the numbers reported by different datasets such as the Penn World Table (PWT) or the World Development Indicators (WDI): First, data providers may base their series on different vintages of sometimes substantially revised National Accounts data, and second, they often account for shortcomings and gaps in the original data using different methodologies. A growing body of literature is concerned with the impact of such differences on macroeconomic inference (e.g., Ciccone and Jarociński, 2010;Ponomareva and Katayama, 2010;Johnson et al, 2013;Breton, 2015;Breton and García, 2016). Two core findings of this literature motivate the study at hand: First, the GDP series tend to diverge significantly more for countries with low incomes.…”

Global trends in relative and absolute income inequality

“…9 Please note that the 2005 PPP estimates have been heavily criticized (e.g., Deaton, 2010;Breton and García, 2015) and that it seems that the newest PPP estimates (from the 2011 round) are similar to the pre-2005 estimates (Deaton and Aten, 2015). However, 2011 PPP data are only available from 1990 onwards.…”