of
Information Technology:
A Review
Home Page
We present in brief the theoretical frameworks, which are being used to address the problem of investments in information technology.
1. Consumer Surplus
2. Production Function
3. Factor Demand Framework
4. Business Performance Matrix
5. Value Based Models
6. Transaction Cost Economics
7. Contingency Theory
8. Theory of Organizational Design
9. Agency Theory
The effect of IT can be examined by measuring the welfare effects (consumer surplus) associated with their use. The assumptions underlying this approach are that decision-makers are rational and able to understand their activities better than others, and that the money which they are demonstrably willing to spend on these technologies is at least a minimum measure of the worth of these technologies. Using this rationale, one can calculate the internal rate of return (IRR) and use it as an indicator of the worthiness of a particular investment. However, due to the externalities associated with these technologies and the possibility that consumers may actually benefit more than they pay for a technology, the IRR is often supplemented by estimates of consumer surplus. Brynjolfsson (1994) used data from the US to estimate the increase in consumer surplus associated with the decline in computer prices during the 1975-1987 period.
Different measures of consumer surplus:
Marshallian surplus: The most common method of estimating consumer surplus is based on the ordinary or the Marshallian demand curve. Given a specification of the demand curve, one can directly evaluate the Marshallian consumer surplus by integrating it between any two prices. Log-linear specification of the demand function can be given as:
Q = eg pa yd
Where, Q is quantity, p is price, y is income and g, a, and d are parameters to be estimated. Integrating the demand curve from the initial price p0 to the final price p1 yields:
CS? = ?eg yd (p11+a - p01+a)/ (1+a)
Given prices, income and estimates of the parameters, consumer surplus can be calculated. However, the Marshallian surplus is not an exact measure of welfare.
Compensated demand curves (Hicksian demand curves): It gives exact surplus based on a demand curve adjusted so as to maintain the same utility level before and after the price change. In the case of a log-linear demand function, the consumer surplus associated with the compensating variation in income (CVI) principle may be calculated as:
CVI = [y1-d + (1-d)(eg)(p11+a - p01+a)/(1+a)]1/(1-d)
The additional terms in this expression compensate for the change in real income due to the price decline. These two methods of calculating consumer surplus are based on the estimation of a parametric demand curve. Imposing such a restriction on the shape of the demand curve may not be appropriate and may lead to misleading estimates.
A non-parametric derivation consumer surplus: The assumption used above that the parameters of a functional form of the demand curve can be estimated may not be always true. While this is generally possible, some error may be introduced in the estimation procedure, especially if the functional form chosen does not fit the actual demand curve well. The alternative non-parametric derivation approach is to explicitly add up each of the additional increments to consumer surplus from each price decline. While this approach is more tedious than directly integrating the whole curve, it makes use of data on intermediate points which may not lie exactly on the equation for the estimated demand curve. The formula used for this approach is given by equation:
S(pt – p1) Dqt (y1 / yt), for t = 0,……..,1
where, pt is the actual price in period t, Dqt is the actual price change between pt-1 and pt, yi is the income in period t
Value based on the theory of index numbers: It is based on the idea that consumer welfare is captured by the increase in utility resulting from changes in prices and consumption. Accordingly, instead of relying on a demand function, one may use a utility function to derive a measure of welfare. Based on a translog utility function and some results from the theory of index numbers, Bresnahan (1986) derives such a measure. The measure of consumer welfare associated with such a utility function may be represented as:
1/ 2(S1 + S0) log (p0/p1)
where, Si ( i = 1,0) is the share of IT in total expenditure in the final and initial period, respectively.