The former head of Procter & Gamble’s U.S. operations, Durk Jager, recently characterized the existing trade promotion system between manufacturers and retailers as “impossibly inefficient.” He may well speak for many manufacturers, and retailers too. A common practice in the packaged goods sector, trade promotion involves manufacturers giving cash or a discount to retailers in hopes that the retailer will in turn discount or promote their product to consumers. The arrangement, also found in the apparel, shoe and electronics industries, is a major source of friction between the two parties.

According to Progressive Grocer magazine, two-thirds of retailers feel they don’t receive a fair share of trade promotion dollars. Yet manufacturers aren’t happy either, with 85% believing that their trade promotion dollars are not being spent effectively. At the same time, the last couple of decades have seen an overall marketing trend away from advertising and toward the more traceable promotions (coupons, discounts and the like), with the lion’s share going to trade rather than consumer promotion. But if trade promotions are so unprofitable, why are they increasing? Why don’t manufacturers stop offering them? “There’s no way to get all the manufacturers together to agree not to do it … As a result, they all end up in the trade promotion trap,” says Wharton marketing professor David Bell.

Bell is the co-author, with Xavier Dreze of UCLA’s Anderson School of Management, of a new paper entitled Pay-for-Performance Trade Promotions: Why They Work and How to Implement Them. In their paper, Bell and Dreze assert they have found a path out of this vicious cycle for both manufacturers and retailers. Their solution lies in a variation on a relatively new type of trade promotion known as pay-for-performance. If manufacturers are willing to sweeten their pay-for-performance deals sufficiently, say the authors, then both they and retailers – and ultimately consumers – will be much better off. Paradoxically, manufacturers will have to “give more to get more”- not always an easy strategy for managers to accept.

In explaining why trade promotion is such a thorn in the side of manufacturers in particular, Bell and Dreze detail some of its unfortunate side effects. In the traditional type of offer, known as an “off-invoice” deal, the manufacturer rewards the retailer based on the number of units purchased over a set period. Retailers, eager to maximize their own profits, have responded by loading up on the product while it’s being offered at a discount, then deliberately not passing the discount on to consumers. Instead, they sell the product later at the regular price (forward-buying). Or they sell it to other retailers for a profit (diverting).

“A manufacturer usually offers a deal in only certain regions of the country. So the smart retailer in Boston might buy much more than he needs for his own store, put the rest on a truck and ship it to a retailer in Philadelphia,” Bell says. Not only does the manufacturer lose out because the promotion never reaches consumers, but the retailer loses out in that he or she must tie up money and effort in warehousing, refrigerating or shipping the excess inventory.

Yet, according to Bell, forward-buying and diverting are so commonplace they have become the rule: “I’ll tell you how bad the problem got: About five or ten years ago, a company actually had a public offering of stock based on a business plan to provide storage facilities for retailers who were diverting.”

In response, manufacturers have in the last few years come up with a new form of deal known as pay-for-promotion, or scan-back. The pay-for-promotion deal introduces a critical new stipulation: Retailers get rewarded only for as much product as they can prove was sold from their particular store in the given deal period. This type of deal has become more feasible with the advent of modern store scanners, which makes it easier to record and verify sales. Pay-for-performance helps eliminate forward-buying, because only the units sold to customers during the promotion period get the price discount; it also helps eliminate diverting because the eligible units must be sold from the retailer’s own store. The authors cite data that off-invoice allowances were down from 50% of the trade promotion budget in 1995 to 33% in 1999, while scan-backs and discretionary funds together were up to 58%. This suggests that moves are afoot to rely less on off-invoice and more on scan-backs, say Bell and Dreze.

Are scan-backs really more cost-effective for the manufacturer? The researchers report findings from a national brand beverage manufacturer that field-tested both off-invoice and scan-back deals, offering them to retailers in four different regions of the U. S. at different times over a year. The results revealed that when using scan-backs rather than off-invoice, this manufacturer experienced much less forward-buying, a larger pass-through of the deal to consumers and more stable retailer demand.

But many retailers are still leery of scan-back deals, which makes perfect sense, say the authors. After all, they’re accustomed to generating sizable profits through forward-buying and diverting. What’s more, scan-back deals require them to disclose their sales figures to the manufacturer. The average retailer, given the choice, will always choose off-invoice over scan-back. The solution, say Bell and Dreze, is to sweeten the scan-back.

“If there’s an ‘aha’ moment in our paper, it’s this: As a manufacturer, you will actually do better by seeming to give away more money,” says Bell. “If I’ve offered you an off-invoice deal of $2 off every $20 case of Tide you order in the next two weeks, the only way I can create a scan-back deal to beat that is to give you perhaps $4 off a case, or lengthen the promotional period to four weeks. The interesting twist is that even if I’m giving you $4 off a case, I’m actually going to be better off because you won’t be encouraged to forward-buy. With well-designed pay-for-performance deals,” Bell continues, “manufacturers can have better assurances that their dollars will reach consumers. Retailers, for their part, can dramatically cut inventory costs and re-orient their activities around what should be their core competencies – selling and marketing.”

While pay-for-performance is an important step in the right direction, the authors say, it isn’t a “silver bullet” for the often-acrimonious relationship between retailers and manufacturers. Bell notes that retailers can (and do) undermine even scan-back deals: They can artificially inflate recorded sales, or in a worst-case scenario, purchase diverted merchandise and sell it through their own store’s scanner during the promotion period, forcing the manufacturer to pay twice. Asserting that trust needs to be built between the two sides, the authors propose introducing a third-party auditor into the relationship. The auditor would fulfill functions such as informing the retailer of the deal terms, verifying sales and paying the retailer.

Bell and Dreze make recommendations for managers on how to choose an auditor. They acknowledge that organizational structures, too, will have to be altered in a pay-for-performance environment. “Clearly, there is some inertia here, and it will take time for retailers to orient their operations and compensation structures around selling rather than buying.” They also assert that manufacturers, mired in administrative tasks, have not scrutinized trade promotions sufficiently with an eye toward change. Overall, the authors argue for collaborative partnerships between manufacturers and retailers to build credibility and understanding.

Will pay-for-promotion eventually dominate, with both retailers and manufacturers realizing – as the authors hold – that greater profits lie in that direction? “In some big-picture sense, retailers are aware that the old off-invoice system creates problems for them – that it causes a lot of inefficiency and adds costs so that everyone is worse off in the long run,” Bell says. “Pay-for-performance is all about compensation based on transparent results. That has to be a good thing.”