Efficiency, equity, and the distributional impacts of development projects

(Note: This reflection paper was written in the first quarter of 2019 as part of a cost-benefit analysis course)

Briefly describe the trade-off between efficiency and equity in Economics. Normally, which one is positive analysis, which one is normative analysis? In Loomis (2011), is the distributional analysis a positive analysis or normative analysis?

Efficiency is concerned with the optimal production and allocation of resources given existing factors of production. For example, producing at the lowest cost. While, equity is concerned with how resources are distributed throughout society. Market inefficiencies are referred to as market failure by economists. Market failures lead to problems like free riding, under consumption of a positive good (for example, vaccination) or over consumption of a negative good (for example, pollution). These concerns are addressed by regulations such as taxation or subsidies to redistribute resources so as to attain a market efficient equilibrium. However, in most cases, the market efficient equilibrium is not equitable, because it makes the rich richer and the poor poorer. In this case, an over emphasis on efficiency leads to an unequitable outcome. Therefore, especially for social sector projects, the demands of efficiency have to be balanced with the need for equity.

Normally, positive analysis is the scientific method; where efficiency is the standard. While, normative analysis is the normative analysis; where concerns of ethics and equity are the yardsticks for measuring success. While the rationale for the distributional analysis is normative, it’s implementation makes it a positive analysis because it systematically tries to measure and assign weights to cost and benefits for various groups affected by the policy or project.

Summarize the Implicit Weighting approach and the Explicit Weighting approach. Provide two empirical examples used to calculate the distributional weights.

The Implicit Weighting Approach analyzes the actual choices and decisions of various stakeholders, and assigns distributional weights based on these. The implicit weighting approach uses cross tabs and regressions to analyze these actions so as to identify specific cost and benefits for various sub-groups. Therefore, the implicit weights are disaggregated project costs, benefits, and net benefits by demographics, user or stakeholder groups. For the implicit weighting approach, costs are usually assigned based on the source of financing of the project; user fees or taxes, while benefits are usually assessed through surveys.

The Explicit Weighting Approach assigns project beneficiaries into groups based on preexisting categories or other justifiable criteria and assigns weights to each group that reflect the marginal utility the group receives from their net benefits. The analyst could further compare the baseline case of equal weights with when explicit weights are applied.

This explicit weight can be assigned using a revealed preference (existing progressive income tax levels as source of weights), relativist (willingness to pay as a percentage of income provide a relative comparison of benefits to each income class), or Lorenz curve (distribution of project benefits across the beneficiaries before and after the project) approach.

Empirical examples used to calculate the distributional weights

I. Using Survey Demographics to Display How NonMarket Benefits Vary by Income: Using Contingent Valuation Method, residents in Fort Collins, Colorado were asked through a mail survey the maximum amount they would pay annually to avoid a 50% reduction in peak summer river flows. While benefits varied randomly across income groups, the costs side revealed that distributional burden would depend on how project was financed. The property tax though not perfect was a better option compared to the sales tax, since housing (property) expenditure rise with income.

II. Hedonic Property Value Analysis of the Effect of Nearby Forest Fires on Hispanic and Low Income Residents Home Prices: Using regression analysis, the impact of forest fires (x) on residential house prices (y) was estimated across various groups. The coefficients for each group informing the distributional weights to be assigned.

Carefully read the Hedonic Pricing Method Example.

· Is it an implicit weighting approach or an explicit weighting approach?

The Hedonic Pricing Method example is an implicit weighting approach because it utilizes values derived from data from surveys to derive the values of weights to be assigned to categories/groups of interest.

· According to the Hedonic pricing model P = func(E, S, N), which variables from Table 2 belong to E, to S, and to N?

a. E are the location specific attributes. In the model, the variables are: Log (distance to fire), Elevation, Distance to USFS Land, and Distance to Los Angeles.

b. S are the house characteristics. In the model, the variables captured for this is House Sq Feet.

c. N are neighborhood social and demographic variables. In the model, the variables are: % Hispanics, and Household income.

· Based on Table 2, interpret the signs of the following 3 coefficients: log(distance to fire), %Hisp*Distance to Fire, Income*Distance to Fire (you can ignore the magnitude).

a. Log (distance to fire) — Log (distance to fire) is associated with an increase in Log House Sale Price — the farther a house is from the fire the higher its’ sale price is (positive coefficient 0.087 (statistically significant — at both 5% and 1% levels)).

b. %Hisp*Distance to Fire — %Hisp*Distance to Fire is associated with a decrease in Log House Sale Price — with the same distance to fire, house prices in neighborhoods with a higher percent Hispanic population are lower than house prices in neighborhoods with a higher percentage of white population (negative coefficient

-0.0000056 (statistically significant — at both 5% and 1% levels)).

c. Income*Distance to Fire — Income*Distance to Fire is associated with a decrease in Log House Sale Price — with the same distance to fire, house prices of low-income individuals are lower than house prices of high-income individuals (negative coefficient -0.000000000577 (statistically significant — at both 5% and 1% levels)).

· Which income group and which minority group benefit more from the forest fire prevention projects? Elaborate why it “might raise distributional concerns depending on how the program were financed”.

High income individuals and white neighborhoods benefitted more from the forest fire prevention projects.

The program could have been financed by user fees or taxes.

If the project was financed by taxes, it would raise distributional concerns because high income individuals and white neighborhoods are benefiting more, except the tax system is progressive and well implemented. These are groups that are already better off initially. This creates more inequality. The marginal utility for these groups is lower. Overall, this leads to a reduction in social welfare.

If the project was financed by user fees, it would be possible to identify which users are benefitting more and get them to pay the value of what they are benefiting from the fire prevention efforts.

In your own opinion, why considering distributional impact should be a key step in CBA?

Because a social sector (development) project should make the society better off, projects benefits should get to the groups who need it most so that the project does not lead to more inequality in the society. Considering distributional impact should be a key step in CBA because a development project is expected to contribute towards maximizing social welfare.