Chapter 2: Getting Axed and Landing in a Halfway House

Read our previous post Origins: How we Became CapTrade —Chapter 1: We Fly the Coop

During the second half of the 1980s, the original four CapTrade principals (Anderson, Klett, McCleary and Rogers), along with vital contributors Peter Kuhbach, Dr. Cindy Lewis, Jack Nightingale, and Dean Shulman, built a small trade regulation consulting practice at the “Big 8” accounting firm of Coopers & Lybrand.  Conflicts with existing C&L clients often prevented the group from undertaking some projects, but the practice was growing rapidly, turning a profit, and serving some of the firm’s largest global clients, including Samsung, Toyota, and Ford. 

One day in early 1988, we arrived at our offices at the corner of 18th and M Streets in downtown DC and found out that C&L was eliminating the larger international trade consulting practice that was our home.  It took a few minutes for the implication of that message to sink in, but we slowly figured out that we were being axed.  The following day, the C&L DC managing partner, no doubt looking at our billings and client list, quickly “un-fired” the trade regulations group and transferred it to the DC litigation support practice.  (That practice fell under accounting, which was ironic, as none of us were accountants!)  Thus, while the sacking had lasted only about twenty-four hours, it was unnerving and spurred us to search for an alternative home for our practice.

Bob Reynolds, the husband of Cindy Lewis, happened to have a thriving economics consulting practice at ICF, a mid-sized consulting firm just down the street from C&L.  Having had an active trade regulations practice in the past, and a large complementary antitrust practice, ICF seemed like a good home for our practice.  ICF’s smaller size was also attractive, as it likely would reduce the number of cases we would have to turn down because of conflicts.  So, after a relatively brief negotiation period, we moved our trade regulations practice to ICF, located at 1875 K Street, NW.

The iconic International Square building, home of ICF, located at 1875 K St NW in DC.

An entrance to the equally-iconic food court, with its own escalator just outside the Farragut West Metro station entrance.

ICF allowed us to choose any computer we wanted.  Most of us went with the latest Compaq portable, which featured an orange plasma screen and a detachable keyboard (Laptops had not yet hit the market).  These were pricy machines. During “down time,” we would load up “Flight Simulator” and see if we could fly the Lear Jet across the country and land it on the Golden Gate bridge.  One unnamed member of our group played so much that he wore out the directional keys of the Compaq keyboard, which had to be replaced to the tune of $400.

One of the more interesting cases we worked on at ICF involved Korean man-made fiber sweaters.  A bit of explanation for those of you who don’t live in the world of antidumping.  As the measurement of dumping essentially is a comparison of prices across national markets, it requires decisions on how to different products that are sold in different markets, in order to compare prices.  If the product is as simple as nitrogen fertilizer, then the product matches are obvious.  But the more complex the product, the more difficult it is to match products, especially (as often is the case), design aesthetics, standards, and even units of measurement differ by market.

One of the best fountains in DC, located inside the atrium of the International Square building.

View of office exteriors, facing the atrium and its skylight ceiling.

Back to sweaters.  The big challenge in that antidumping case was to come up with a way to match the specific sweater designs sold in the U.S. market to those that were sold in Korea.   On the fly, we had to immerse ourselves in the intricacies of knitting design and production – stitch types, neck styles, sleeve styles, yarn types, accessories, and the like.  (One nice thing about spending a career in dumping is that you gain a lot of knowledge about many different products, which you can use to impress friends and acquaintances at parties!)  The U.S. Department of Commerce lacked a systematic methodology for matching products across markets in dumping cases.  Commerce would send out a letter in each case asking the parties to identify the best home market product match for each product sold in the U.S.  This approach had some limitations; namely, that Commerce rarely considered the second- or third-most similar product in its dumping calculations.  If the proposed matching product fell out of the dumping calculation program because it was below cost, Commerce went straight to constructed value. 

Back to sweaters again.  With an infinite number of combinations of design, materials, accessories, color, stitch type, patterns, and sizes, we were worried that there would be no way to defend any challenges to our proposed product matches.  Peter Kuhbach, a brilliant programmer and a member our team, came up with the idea of establishing a hierarchy of physical characteristics, assigning each unique characteristic a numeric value, concatenating the numerical values of each product in the home market and the U.S., and then finding the “most similar” product by subtracting the concatenated numerical values from each other.[1]  Peter would spend hours with the latest laptop computer, perfecting his “model matching” program, to get it to run in the shortest amount of time.  One of his contributions was to label the concatenation the “Control Number.”  Quickly abbreviated to CONNUM, this acronym has remained in the antidumping lexicon ever since.

Our team now had a systematic method for coming up with product matches.  It worked so well that, after the sweaters case was over, the Commerce official in charge of the Department’s SAS calculation team called us and “invited” us to turn over the code, as it was exactly what Commerce needed to comply with a recent court ruling that Commerce give more priority to prices in antidumping calculations, rather than going straight to constructed value when the “best match” dropped below cost. 

Another big development while we were at ICF was the introduction of a PC version of SAS, the computer program used by Commerce to calculate dumping margins.  Prior to PC SAS, Mary Ann McCleary and our other programmers would have to go off-site to run the dumping margin code.  And finally, the courts ruled that Commerce would have to release its dumping program.  Now, we really were able to estimate potential duties – for both sides of the investigations!  The early computer days were a bit more stressful.  Data downloads could take a full day.  For submission to Commerce, the data had to be transferred to reels of magnetic tape, a tedious exercise that sometimes took hours.

The ICF we joined in 1988 was a growing and privately held consulting firm, focused on the environmental market.  Not long after our joining the firm, however, ICF went public.  The funds raised with the IPO were used to acquire a number of other consulting firms in a broad range of areas.  In our second year there, right after ICF informed us that a large portion of our annual bonuses would be paid in stock options, the company announced massive losses (largely associated with those new consulting practices they had bought) and promised the market that they would refocus on their “core” practice areas.  Rather than wait to hear once again that we were no longer wanted, we decided to move on.

But where would we go?

Next up:  Chapter 3:  The Accidental Birth of Capital Trade

[1] See Valerie Owenby’s explanation on “most similar” product matching, “The ‘Most Similar’ Penguin.”