Overstock ordered to pay multimillion penalty for falsely advertising comparison prices

People v. Overstock.Com, Inc., --- Cal.Rptr.3d ----, 2017 WL 2391814, No. A141613 (Cal. Ct. App. Jun. 2, 2017)

The trial court found that Overstock, an online retailer whose value proposition is “the lowest prices on the Internet,” had engaged in unfair business practices and false advertising; it granted injunctive relief and imposed $6,828,000 in civil penalties. The court of appeals affirmed.  One lesson here: there is a lot of documentary evidence, mainly in emails to suppliers and internal emails, that Overstock knew it was manipulating price claims; compliance didn’t seem to become a priority in the company until far too late.

Overstock started mostly with products from businesses that were liquidating excess inventory, but now gets most of its goods from third party “fulfillment partners.”  Its product pages used various terms to compare prices.  At the beginning of the relevant period, it showed a “List Price” for the product, with the number stricken through, then Overstock’s price, then a calculation of the difference in dollars and percentages.  It eventually changed “List Price” to “Compare at” and then to “Compare.” A company executive testified that advertisement of reference prices gave customers confidence that they were shopping at a site that offered real savings. An employee email stated that “compare at” pricing “definitely helps entice the customer to purchase.” Products that did not have “compare at” prices suffered reduced sales.  (From 70-90% of Overstock’s products weren’t at issue here, because they had a list price derived from standard industry data—these were mostly books, movies, music, and games.)

Before late 2008, Overstock had no process in place to ensure that all comparison prices were verified. Its policies allowed the list price to be set by finding the highest price for which an item was sold in the marketplace, but Overstock didn’t consider whether other Internet retailers had made any substantial sales at the comparison price. An Overstock manager told employees, “I need you to find the HIGHEST selling price. We found out it can include freight, which will make it even higher.”  Another email: “Oh, I think it’s been established that the ‘List Price’ is egregiously overstated. This place has got some balls.” Overstock at least sometimes asked a supplier to raise its own website price to make Overstock’s prices look better or to “bump up” the manufacturer’s suggested retail price (MSRP).  Overstock also sometimes used “formulas” to derive list prices, such as doubling or tripling the cost to Overstock or the usual wholesale cost.

In 2007, a Shasta County resident bought two identical patio sets that showed a list price of $999 and an Overstock price of $449. He believed that, because of that supposed retail price, the sets would be of good quality, but they weren’t. One of the tables had a sticker showing it was sold by Wal-Mart for $247. Although he received a full refund, the patio set remained on Overstock’s web site, with the same purported list price, for “quite a while.” He wasn’t the only one to complain; he also reported the matter to the district attorney.

In response to the resultant investigation, Overstock sent a letter to its fulfillment partners about acceptable ways to set a “list price.”  Then Overstock changed its term from “List Price” to “Compare At,” but for a year there was no change in Overstock’s policy on how comparison prices were set and Overstock did not have a process in place to verify that a product had been sold at the comparison price. Overstock employees still discussed with suppliers the possibility of raising their MSRP so that Overstock could show a higher discount and Overstock continued to use comparison prices that were the highest price at which an item was offered for sale.  Among a random selection of ten of the top 100 selling products from each Overstock department in 2008, “compare at” prices were on average 15% higher than the highest actual selling price on line. Among 10 products from the top 100 items with the greatest “you save” percentage, the “compare at” prices were 33% higher, and among a random selection of ten additional products from each department, the “compare at” prices were 13% higher. 

Overstock then removed the “compare at” pricing from most of its products and allowed them to be re-posted only if the fulfillment partner provided a verified reference price. Sales of products without a “compare at” price dropped 6% or more; conversion increased 9% when “compare at” prices were added. Overstock also formed a “pricing validation team” to verify that the items it sold were actually sold elsewhere at “compare at” prices, and to re-verify those prices every 90 days. The team sought to verify the highest street price for the product. There were still instances in which Overstock discussed the possibility of the partners raising the prices for their products on their own websites or on Amazon in order to create a higher comparison price. The team sometimes verified a high street price for an item that was similar, rather than identical, to that sold on Overstock.

Overstock received numerous complaints from customers that the “compare at” prices were inflated.

The People’s expert testified at trial that, as advertised reference prices increase, the price that a consumer thinks something costs, perception of product quality, and the “perceived value” of the product all increase, which makes consumers less likely to continue comparison shopping and more likely to buy.  Although consumers often discount reference price claims, the claims still affect their perceptions, and many don’t think the reference prices are inflated. Overstock’s expert concluded that a majority of Overstock’s customers didn’t notice or care about reference prices. However, he acknowledged that ARP’s can increase customer loyalty and that between 70-75% of participants in a survey he conducted believed such terms as “MSRP,” “compare,” and “compare at” reflected “regular average” prices.

Along with rejecting Overstock’s challenge to the application of a four-year statute of limitations, the court of appeals also held that substantial evidence supported the trial court’s factual findings that it violated the FAL and the UCL.

Overstock argued that its list prices were “fair estimate[s]” insofar as they were based on markups that are generally used to set retail prices over wholesale, and that the trial court had no reason to conclude that “List Price” was more than an estimate.  But on the face of the advertising itself, the trial court’s conclusion that “List Price” means “list price”—which Webster’s defines as “the basic price of an item as published in a catalog, price list, or advertisement”—required no other logical or evidentiary support.

The trial court also ruled that Overstock knowingly misled consumers when it used prices from similar products and formulas to set some reference prices during the time that it used the terms “Compare at” and “Compare,” unaccompanied by qualifiers such as “similar.”  Experimental and survey results indicated that, “with respect to any particular nomenclature, a significant portion of consumers view the given label as reflecting a ‘regular/average price’” for the same product.  Overstock argued that the People didn’t submit substantial evidence that consumers were actually misled, but that’s not the standard: likely deception is. 

The misleading character of these terms appeared on their faces. “Specifically, the use of strikethrough font (‘Compare at $190.00’) followed by ‘today’s price’ and then a precise calculation of the purported savings, clearly suggests the actual item’s price has been reduced.”  Plus, the evidence showed that Overstock’s practice invited other abuses, such as using items that weren’t truly similar. In one case, an Overstock email said: “Comparing this dress to a DVF dress is like comparing a Lexus to a Geo Metro.” Another email to a fulfillment partner said: “Just make sure similar size and materials [are used for comparison]. The higher [the price] the better.”  The FTC’s Guides Against Deceptive Pricing also supported this conclusion, since it requires that comparisons be to merchandise “of essentially similar quality and obtainable in the area.”

There was also sufficient evidence that Overstock’s use of formulas was misleading. Overstock manipulated markups to achieve a specific amount of “savings” for the consumer, e.g., “If the item is a true close out and can’t be found anywhere online I will make sure the MSRP listed provides a 40 to 60 percent discount.” “Advertisements showing a ‘compare at’ price followed by a calculation of savings down to the penny, without informing consumers they are looking at an estimated (and possibly manipulated) price, manifestly had the capacity to mislead.”

Likewise it was misleading for Overstock to set ARP’s “based on the highest price that can be found without regard to the prevailing market price and without any disclosure of the practice.”  One customer testified that she was “outraged” to learn that Overstock used the highest available price, which she called “lying.”  The FTC’s guide likewise wanted realistic comparison prices.  The People’s expert also testified about a study that showed 72% of participants thought such terms as “regular price,” “compare at” or “manufacturer’s suggested list price” represented either the price an item usually sold at or its price at most other stores, rather than a fictitious, inflated price. Overstock’s own survey, conducted by a third party, showed that more than half of respondents wanted price comparisons to reflect the “Average retail price,” and Overstock’s own expert agreed that a substantial majority of consumers thought that the terms at issue represented “regular average” prices.  Many actual customers complained that the “Compare at” prices were inflated.

The FAL applies to an untrue or misleading statement “which is known, or which by the exercise of reasonable care should be known, to be untrue or misleading,” and Overstock argued that there wasn’t sufficient evidence that it knew its practices would mislead consumers.  When formulas or “similar” products were used during the “List Price Era,” Overstock knew the term “List Price” was false.  As for the rest, the same evidence that supported misleadingness strongly supported an inference that Overstock knew or should have known of this likely impact. 

Overstock argued that its own hyperlinked disclosures showed that Overstock wasn’t knowingly misleading. But Overstock’s own disclosures about how it defined “compare” proved nothing about what the advertisements would reasonably be expected to convey to the consumer. (Those disclosures said “List price is not necessarily the lowest price at which the product is commonly sold, and often will be higher than the actual price at which the product is sold.” Overstock conceded that its disclosures “may not have been prominent enough” to affect consumer understanding.)

Proof of knowledge can be circumstantial. The “you save” calculation in the advertisement would be useful only if consumers perceived the ARP as a “real” price. “That this was consumers’ perception—and that Overstock knew this—is demonstrated by the irate communications to Overstock when customers discovered the ARP’s were not real, but inflated prices,” e.g.: “[T]his is not the truth since there is NO place that this camera truly sells for $250.77.... You are advertising lies. This is NOT an incredible savings. The market price of this camera is $99.99, NOT $250.77.”

Penalties: Both the UCL and the FAL authorize civil penalties of up to $2,500 for each violation:

In assessing the amount of the civil penalty, the court shall consider any one or more of the relevant circumstances presented by any of the parties to the case, including, but not limited to, the following: the nature and seriousness of the misconduct, the number of violations, the persistence of the misconduct, the length of time over which the misconduct occurred, the willfulness of the defendant’s misconduct, and the defendant’s assets, liabilities, and net worth.

The trial court did not measure the number of violations by the number of Californians who saw the offending advertisements or by the number of sales made through the offending pages, which would both have resulted in penalties of hundreds of millions of dollars.  Instead, the trial court used the number of days Overstock violated the statutes, with a daily penalty of $3,500, calculated as $1,000 for each of the three types of violations (basing ARP’s on formulas, nonidentical products, and the highest possible price), with an additional $500 for “the lack of controls that led to various abuses.”  From the date the validation team process was implemented and Overstock no longer used formulas, the daily penalty dropped to $2,000.

The trial court found the “ ‘seriousness of the misconduct’ ” was moderate in that the misconduct was less egregious than that in other reported cases, Overstock’s prices were at or below those of its competitors, and the misconduct affected only some of Overstock’s product lines. But the offending practices were numerous and persistent, on thousands of product pages. The court found that Overstock’s conduct was willful, in that it was inconsistent with the guidelines of the FTC and the BBB. The trial court also noted that it had declined to order restitution to customers because of the difficulty of identifying an appropriate award or appropriate recipients, and that this lack of restitution was a factor in setting the appropriate penalty, which was the minimum necessary to fulfill the statutory purpose.

Overstock argued that this penalty was an abuse of the court’s discretion; the court of appeals disagreed. Among other things, the court rejected Overstock’s argument that there was no “concrete injury” to consumers. “The fact that Overstock in fact (according to its undisputed evidence) offered the lowest prices in the market does not mean no injury occurred.”  The false ads decreased consumers’ search intentions, increased their perception of the transaction’s value, and increased the likelihood that they’d return.  There was no restitution awarded because it was too hard to calculate, not because there was no harm.  Although the penalties here were larger than those upheld in any published case, that didn’t make them excessive.

Nor did Overstock’s “it’s the government’s fault because they let us get away with it for so long” argument succeed.  The People didn’t sue until more than three years after the Shasta County DA began investigating it, and it took an additional three years for the matter to proceed to trial, allowing penalties to accumulate. But Overstock waived this issue by never raising it in the trial court.  Even if it weren’t waived, it was silly; Overstock was aware of the investigation since 2007 and negotiated a tolling agreement to delay the start of litigation. “In the meantime, it chose to continue the offending practices.”

Nor was the penalty so grossly disproportionate to the gravity of its offense that it violated the Eighth Amendment. “The penalty the court set was both far below the maximum allowed by statute and well within Overstock’s ability to pay.”

As for injunctive relief, “‘The remedial power granted under these sections is extraordinarily broad. Probably because false advertising and unfair business practices can take many forms, the Legislature has given the courts the power to fashion remedies to prevent their “use or employment” in whatever context they may occur.’”  Although the issue was close, the trial court didn’t abuse its discretion in enjoining the use of formulas, which Overstock had abandoned in 2008. At and after trial, Overstock continued to take the position that the use of formulas was proper, and even after ceasing the use of formulas, it continued to use other misleading practices.  Given the record evidence about Overstock’s use of inflated MSRPs, it could also be enjoined from unqualified use of the term, and it could be required to re-validate every 90 days. The fact that it started doing so voluntarily, after litigation began, didn’t make an injunction inappropriate.