I would like to thank Evelina Tainer, chief economist here at Econoday, for asking that I do a "Short Take" on a new retail sales survey that I've developed. I promise to keep exaggeration to a bare minimum.

In January this year a new weekly retail sales report appeared on Market News International (MNI). The report, developed by yours truly in cooperation with Market News, tracks company statements from the U.S. retail trade sector. Revenue and profit data are compiled and industry comments and outlooks are compressed and highlighted. In a unique twist, the data are assessed in a diffusion format that separates results into three categories: better, same, worse. An index emerges, one based on the "50-breakeven" equation used by purchasing surveys.

Five years of heightened regulatory oversight have improved the quality and increased the flow of corporate information. In no industry is the volume of this information greater than retail trade, where many of the groups -- discounters, department stores, apparel chains, restaurant chains -- routinely issue monthly sales results and guidance. Retailers also issue a mix of mid-quarter updates, end-of-quarter statements, and industry presentations. This stream multiplied by the more than 300 retail chains that are publicly traded in the United States make up the survey's pool of data.

At a reading of 50.2 in the four-week period ended May 21, MNI's Retail Trade Index is currently pointing to steady but slow conditions in May. General merchandise (representing nearly 54,000 individual U.S. stores) is a leading group with apparel (6,500 stores) and restaurants (13,000) lagging. High-end chains are reporting strong results in contrast to close-out chains that are reporting weakening results and diminishing traffic. Typical comments include those from Stage Stores (537 family apparel stores - STGS) which said it is pleased with the pace of business so far in May and pegged the full month's gain in the mid-single digits. Harvey Electronics (8 electronics stores - HRVE) said a resurgence that began in April extended through the first two weeks of May, though it warned that sales of flat-panel TVs continue to be hurt by expectations of price deflation.

The last shall be first
Diffusion measurements are unique in that they are very democratic, treating all input alike. The smallest companies are given the same weighting as the largest. What emerges is a measure that concentrates on the edge, that is on the depth or saturation of change. So far I have found little difference between the performance of large companies and their smaller rivals. This is a large untapped pool of often very candid information. Note that many of these companies, too small to issue corporate debt, are not covered by Wall Street investment houses. Note also the survey excludes all second-party commentary or recommendations. Only statements from the companies themselves are admitted as evidence.

The survey produces a variety of information, headlined by an overall index. I think of it as a barometer of the public record, an index of the retail sector that should be read as a snapshot of current conditions. The index rolls from week to week in four-week periods. Determination of each company's business conditions is based on revenue growth and income growth, both present and projected. Company comments are added into the mix. I put myself in the position of a purchasing manager who is answering a similar survey in the mail, judging simply whether current business conditions are better, the same, or worse than a month ago. An index reading over 50 indicates simply that more companies are reporting stronger conditions than weak conditions, and vice versa under 50.

The graph above compares MNI's Retail Trade Index vs. the widely watched weekly index reported by the International Council of Shopping Centers-UBS. The ICSC-UBS index, which is based on a survey of about 75 chains, showed more weakness at the beginning of the year and more strength of late. Several years of data will have to be compiled before seasonal adjustments can be applied to MNI's data.

A median same-store sales change is also included, as are income data. Income data play an important role in the determination of business conditions, a factor that must be kept in mind when looking at economic forecasts. For instance, forecasters may be calling for a big jump in retail sales, but a jump, spurred by a reduction in income margins, can actually lead to declining business conditions. Income data, as well as the roundups of comments and the breakdown of specific sectors, give the report a depth that should be helpful in assessing retail conditions and changes in conditions.

Bottom line
MNI's Retail Trade Index is still in its infancy but I hope it will broaden the look into the retail sector and point to a rich area of research. The data are released the first business day of each week at 10:30 a.m. ET. MNI also publishes a weekly capital-goods survey, developed 2-1/2 years ago along the same principles. Special thanks to Market News International.

Mark Pender, Contributing Editor, Econoday