Dr. Muttukrishna Sarvananthan, Principal Researcher at Point Pedro Institute of Development presented some of his work on research methods for contraband trade and on general methodologies for estimating the size of trade in grey/ informal markets.

Colloquium: Methodologies for estimating grey/informal markets – Power point presentation slides

Some points from the discussion are below:

Muttukrishna Sarvanathan (MS): official data is available for goods from Central Bank. For services, data is not as widely available. Contraband trade includes false invoicing of exports and imports (partial evasion of tariffs) and complete non-recording of trade (total evasion of tariffs)

false invoicing can be detected / estimated by:

Partner-country data comparison model and unit price comparison model. There are limitations with both. With the unit price comparison model for example it doesn’t’ account for quality differences (worse in the case of services); capital goods may not be standard; due to market imperfections international prices may be different.

When trying to estimate the amount you can never get a perfect estimate and have to make allowances for various things. You can estimate using survey method or Delphi technique.

Surveys

– The issue with surveys is that you need to build a level of trust with respondents (not a random sample).

– You would have to estimate the amount of goods that are being smuggled by air/sea, work out # of passengers on a particular route, & estimate the number who travel for ‘trade’. get est’d value of goods from traders. Depends on the countries studied, and the type of goods.

Delphi technique

1) purpose sampling. survey method used in management sciences to estimate commercial activity, as some companies do not release data. used by market researchers. based on informed judgments (Chaudhari, 1995:6). qualitative & quantitative data.

2) formal (structured) questionnaire, interview stakeholders (e.g operator, regulator, consumers, etc), 2 or 3 rounds.

3) compare different estimates from different stakeholders. if they differ (most likely) then do second round, using 1^st estimation as a base. estimates should converge after a few rounds.

Telecom sector:

Sriganesh Lokanathan (SL): grey market in telecom includes bypass (services), illegal handsets (goods), etc. To estimate this, perhaps looking at it from the angle of taxes.

Rohan Samarajiva (RS): We’re looking at it from the value chain; how many people are employed b/c of telecom, for example? how much tax revenue is generated from the sector? how much revenue is generated in the BPO industry, etc.. We want to estimate these ‘undefinable’ benefits, not contraband trade. Methodology has to work in all Asian countries.

Divakar Goswami (DG): Delphi methodology may be a good way to work our BPO revenues, for e.g.

MS: BPO industry is booming b/c of cheap labor & transfer pricing

Luxman Siriwardena (LS): we don’t need to know how much contraband trade is taking place, but WHY its taking place. (?)

MS: through Delphi method can obtain qualitative data on reasons also.

RS: if we can estimate the amount of tax generated by the sector, & how much is being traded illegally…..,

LS: estimations of contraband trade will only fuel the government’s slapping tax on medium size enterprises, harming them. danger is greater than the benefits of estimating the informal sector. look at the reasons only.

MS: this is one of the dangers of such research

DG: the more you suppress, the more you have to go underground; smuggling will increase.

RS: need data to make arguments.

DG: the data we want to record is not illegal, so its not a problem.

LS: in the past, taxes were put on cocoa (chocolate powder) resulting gin loss of revenue and fall in demand. local producers could not sell. …..

RS: serivces are 50% of SL’s economy. services can be traded, mode 1 is via telecom. how do we estimate the size of modes1-4 trade?

MS: services are invisible trade, and naturally difficult to measure.

Harsha de Silva (HdeS): at the end of the day we want to know what it takes to increase accessibility. have to be able to identify this.