One of the common criticism on Philippine computation of Consumer Price Index (CPI) is the formula used, the weighted Laspeyres Price Index. There were suggestions even before to change the computation methodology just as United States and other countries change their own methodology from Laspeyres indices to a two stage CPI computation using Laspeyres in the primary stage and then combined these results to create a number of aggregate indices, including the CPI for the second stage of computation (Wikipedia.com). This change in the methodology was driven with the perceived notion that the index systematically overstates inflation because it does not take into account changes in the quantities consumed that may occur as a response to price change.
The same conclusion was made by von Auer (2002) where he concluded in his paper that the price index of Laspeyres is unable to pass inversion test or factor reversal test which gave rise to the idea that the formula tends to measure “spurious inflation” which renders it useless and fallacious.
However, von der Lippe (2008) pointed out that when it comes to the issue of substitution bias, which is one of the major flaws of Laspeyres price index, he commented that in all those cases in which quantities of different commodities will be different to changes over time, there is no basis for interfering price levels from expenditure levels only. Moreover, for his central point to make against von Auer (2002) conclustion, when analysing the price movement, the concomitant quantity movement should be taken in to account. A “spurious inflation” of the price index P should not be seen in isolation without realising that there will always be a corresponding “spurious real growth” displayed by the quantity index Q pertains to P. Further, in the case of the “inversion test” or factor test where values remain constant, a price movement is unavoidably followed by a quantity movement in the opposite direction. With the introduction of expenditure weights in the original Laspeyres formula to form a weighted Laspeyres Price index, then the notion of spurious inflation loses much of its charm one the quantity dimension is considered with the conjunction with the price dimension that is once we consider quantities in addition to its prices. This maybe the reason why National Statistical Coordination Board (NSCB) still using the Laspeyres index formula even though some of the countries are now adopting much improve and close to economic theoretic index results.
Bayangos and Estigoy, 2010 study of price index using Philippine data, on the other hand, shows that there is positive bias estimate in the current official CPI measure compared to a geometric CPI (the study used Fischer and Tornqvist Price Index Number for comparison). In particular, the year-on-year growth of CPI using a geometric mean formula is relatively lower compared to that for CPI using a Laspeyres formula by about 0.1 percentage point. They added that the bias estimate may be lower as to the real bias estimates since the result does not yet fully accounted the substitution bias of which the current price index used is liable for. Further, they conclude that this positive bias percentage point indicates that the current CPI measure may overstate “true” inflation.
Another thing that we need to point out, is the selection of base year for the Laspeyres index formula. In the country adopted methodology, there is no fixed interval (number of years) in the rebasing of the CPI (1996, 1978, 1988, 1994, 2000) and solely depends on the availability of the data. The Philippine CPI is reweighted only every six years. And because the CPI is fixed-weighted, it does not adequately reflect on-going changes in the buying habits.
This may greatly affect the computed index number since Laspeyres relies heavily with the base year price and any mistakes made with the choice of base year could result to detrimental outcomes. In particular, a re-weighting of the CPI in a fixed interval basis is necessary to ensure that CPI truly represents the current domestic price condition. Consumer tastes and technology change over time, causing the fixed market basket of goods and services to become outmoded. To capture such changes for a more meaningful price comparison, revision or updating of the fixed market basket, the sample outlets, the weights and the base year have to be undertaken periodically. Thus, the need for more regular conduct of FIES which can be used to reweight the CPI is highly suggested.
Von der Lippe (2008) may pointed out that there are still reasons to believe that Laspeyres could still be a good index formula for computing CPI’s but without proper adjustments made by statisticians, the formula used could not be put to blame.
References:
Bayangos, Veronica B. and I.T. Estigoy. 2010. A Geometric Price Index for the Philippines: A Preliminary Assessment. BSP Working Paper Series No. 2010-01 11
Ericta, Carmelita N. and R. B. Sta. Ana. 2009. Price Collection for the Consumer Price Index: Documentation. Philippine Institutes of Development Studies. Discussion Paper Series no. 2009-19.
T. Srivastava, ShailajaRego 2008, Statistic for Management, McGraw-Hill, New York.
'United States Consumer Price Index', Wikipedia, The Free Encyclopaedia, viewed 20 January 2012, < http://en.wikipedia.org/wiki/United_States_Consumer_Price_Index>
von der Lippe, Peter 2008, Index Formula of Laspeyeres and the Inversion Test, viewed 15 January 2012, <mpra.ub.uni-muenchen.de/7045/>.
von Auer, Ludwig, Spurious Inflation: The Legacy of Laspeyres and others., Quarterly Review of Economic and Finance, 42 (2002), pp.529-542.