9-117-038

R E V : J A N U A R Y 2 2 , 2 0 1 9

Professors Gerardo Pérez Cavazos and Suraj Srinivasan and Case Researcher Monica Baraldi (Case Research & Writing Group) prepared this case with the assistance of Teaching Fellow Iris Leung and Research Associate Quinn Pitcher. It was reviewed and approved before publication by Marc Cohodes. Funding for the development of this case was provided by Harvard Business School and not by the company. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management. Copyright © 2017, 2018, 2019 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800- 545-7685, write Harvard Business School Publishing, Boston, MA 02163, or go to www.hbsp.harvard.edu. This publication may not be digitized, photocopied, or otherwise reproduced, posted, or transmitted, without the permission of Harvard Business School.

G E R A R D O P É R E Z C A V A Z O S

S U R A J S R I N I V A S A N

M O N I C A B A R A L D I

Signet Jewelers: Assessing Customer Financing Risk

“I think they’re heading for a cliff,” said Marc Cohodes referring to his latest short-sell target, Signet Jewelers (Signet).1 Cohodes had made a long career as a canny short seller, with successful shorts on companies from pinball manufacturers to speech-recognition software companies to subprime lenders.2 In early 2016, he had set his sights on Signet, the parent company of jewelry brands such as Kay, Zales, and Jared. Cohodes believed that Signet had become addicted to boosting sales through a risky customer credit program and had a product portfolio consisting of low-quality jewelry— ”trinkets,” as he called them.3 According to Cohodes, Signet was also masking the quality of its credit program through a practice called recency accounting, which allowed them to downplay the number of customers who were delinquent on repayment.

Signet pushed back, with CEO Mark Light decrying the “targeted attack” of Cohodes and other short sellers and affirming that its in-house customer financing program, which the company considered a major competitive advantage, followed “strict risk tolerance standards.”4,5 The company also announced strategic moves, including the potential sale of its credit portfolio and an investment by a private equity firm. Investors and analysts appeared to buy Signet’s argument: its stock price had stabilized by February 2017, buoyed by bullish arguments for Signet’s growth prospects and competitive advantages in a fragmented jewelry industry. 6,7

Cohodes was undeterred. He said, “Their credit book, to me, is beyond toxic. So you have a toxic business, a toxic combination, and I don’t know their way out. I do not know their way out.”8

Signet Jewelers

Company History

Signet, initially known as Ratner Group, was founded in the U.K. in 1949 and rapidly expanded through a series of acquisitions in the 1980s and early 1990s.9 The company rebranded as Signet in 1993 and grew throughout the 2000s both organically and through acquisitions. In 2008, the company

For the exclusive use of J. Vega, 2024.

This document is authorized for use only by Jose Vega in ACG 6175 – Spring 2024 taught by Pietro Bianchi, Florida International University from Mar 2024 to Apr 2024.

117-038 Signet Jewelers: Assessing Customer Financing Risk

2

redomiciled in Bermuda, keeping its headquarters in Akron, Ohio. In May 2014, Signet acquired Zale Corporation, a large competitor that owned Zales jewelry stores, for $1.4 billion.10 By 2016, Signet was the world’s largest retail jeweler, focusing on the mid-market segment of the sector. The company had a 13-14% share of the $41 billion U.S. mid-market jewelry segment. No other competitor had more than a 1% share of the segment.11 (See Exhibit 1 for Signet’s stock price data.)

Brand Strategy

In 2016, Signet employed around 30,000 people and operated in 3,625 in-mall and off-mall stores and kiosks in the US, Canada, and the UK. Signet’s goal was to be the leader in the mid-market jewelry segment wherever it operated. Mid-market jewelry purchases generally ranged from $100-$10,000. Bridal jewelry was crucial to Signet, owing to the large and stable market for jewelry pieces such as engagement rings (see Exhibit 2 on Signet’s merchandise and segment mix and Exhibit 3 for Signet’s financial statements). For Signet’s three main American brands, Kay, Jared, and Zales, engagement rings between $1,000-5,000 comprised 65%, 78%, and 45% of their total stock of engagement rings, respectively.12,13,14 Signet leveraged its large size to buy many of its goods directly from international vendors, a cost advantage not available to the myriad of small players in the jewelry sector.15

Signet used a multi-brand strategy to capture a greater share of the mid-market jewelry sector. The company operated its Kay and Jared stores under the umbrella of Sterling Jewelers. Kay was the largest specialty retail jewelry store brand in the U.S. based on sales. Jared operated stores at free-standing sites across the U.S. Zales operated predominantly in shopping malls, emphasizing diamond jewelry to shoppers who sought jewelry from recognizable designers and brands. Amongst its international brands, Signet owned market leaders H. Samuel and Ernest Jones in the U.K. and Peoples in Canada. Accordingly, Signet’s operations were organized into five business segments: Sterling Jewelers, U.K. Jewelry, Zale Jewelry, Piercing Pagoda, and Others.16

Customer Finance

Customers at Kay and Jared had the opportunity to access a financing program provided directly by Signet.17 Zales stores did not offer the finance program, although they did offer third-party financing through a Zales-branded credit card and planned to adopt the credit program over time.

Signet maintained that its financing program was integral to maintaining its competitive advantage and that no other competitor had the scale and systems necessary to manage an in-house financing program.18 Financing programs required infrastructure to approve loans, service existing loans, and collect in instances where full repayment was not received. Such infrastructure was not cheap: based on the scale of Signet operations, equity research analysts estimated that the SG&A expenses associated with retail financing businesses could total up to 30% of interest income.19

The financing program allowed Signet to reach middle income customers that didn’t have the financial ability to purchase a piece of jewelry outright. This was a boon to bridal jewelry sales, as 75% of these sales in the Sterling division utilized financing.20 In its 2016 annual report, Signet stated,

Our in-house consumer financing program provides Signet with a competitive advantage through the enabling of incremental profitable sales that would not occur without a consumer financing program. Several factors inherent in the U.S. jewelry business support the circumstances through which Signet is uniquely positioned to generate profitable incremental business through its consumer financing program. These factors include a high average transaction value; a significant population of customers

For the exclusive use of J. Vega, 2024.

This document is authorized for use only by Jose Vega in ACG 6175 – Spring 2024 taught by Pietro Bianchi, Florida International University from Mar 2024 to Apr 2024.

Signet Jewelers: Assessing Customer Financing Risk 117-038

3

seeking to finance merchandise primarily in the bridal category; and the minimum scale necessary to administer credit programs efficiently.21

Signet claimed that the “lifetime value” of a customer obtained through the financing program was “estimated to be 3.5 times that” of customers not obtained through the program. Higher average value for in-house financed transactions and high demand for financing in the bridal category drove this additional value. Customer financing also provided the opportunity to generate new revenues from finance charges and fees. On average, Signet’s receivable portfolio turned over every nine months. At the end of January 2016 and 2015, 52.7% and 50.5% of balances due, respectively, were from customers who had enrolled in the financing program more than 12 months prior to their most recent purchase.22 (See Exhibit 4 on Signet’s customers financing statistics.)

Credit applications originating at Signet’s retail locations were automatically approved or denied by a statistical model designed by Signet’s Risk Management team. The algorithm, which took the approval decision out of the hands of commissions-based salespeople, considered credit bureau information, income, employment, address verification, and debt levels.23 Signet also relied on the Fair Isaac Corporation (FICO) score, a credit risk metric that was widely used in the consumer finance industry to assess credit worthiness.24 Signet reported that, from 2014 to 2016, the balance-weighted FICO score for customers utilizing financing was consistently around 660.25 In 2015, individuals earning less than 50% of median family income (MFI) in the United States had an average credit score of 664, compared to 775 for individuals with an income greater than 120% of MFI.26 Overall, the average FICO score for adult Americans in 2015 was approximately 695.27

Grant’s Interest Rate Observer, a leading publication covering debt markets, found that the weighted FICO score of Signet customers was “marginally higher than the 640 threshold of subprime.”28 Individuals with subprime credit scores generally received worse financing terms, generally in the form of higher interest rates.29 Credit card issuers such as JPMorgan Chase and Citigroup typically offered interest rates that ranged from 14% for good and excellent borrowers to 25% for borrowers with average or subprime credit quality.30

Signet reported the performance of its accounts receivable portfolio using the recency accounting method, which required that customers paid at least 75% of their due amounts to remain current (see Exhibit 5 for Signet’s recency accounting practices). This contrasted with the more conservative “contractual method”, which required customers to pay 100% of the periodical obligation by the negotiated deadline to keep a loan current.31 (See Exhibit 6 for opinions on the use of recency and contractual accounting).

Jewelry Industry In 2016, sales in the U.S. jewelry sector reached $61.8 billion. Diamonds accounted for 36% of sales,

although their share of total sales was falling due to lower demand and lower prices. Sector sales grew 13.5% from 2011 to 2016. Sales were categorized into two groups: costume jewelry and fine jewelry. Costume jewelry was generally inexpensive and made with imitation gems, making up 83% of sales volume but only 17% of sales value. In 2016, the average price of costume jewelry was approximately $13, fine jewelry was $311, and luxury jewelry was $1,520.32

The average US household spent $612 per year on fine jewelry and watches. Household income was a major driver of jewelry purchasing behavior: households with annual income of over $150,000 spent an average of approximately $2,000 on jewelry per year. Individuals aged 25-34 years and 55-64 years spent a greater amount than average on jewelry. For the 25-34 age group, marriage-related purchases

For the exclusive use of J. Vega, 2024.

This document is authorized for use only by Jose Vega in ACG 6175 – Spring 2024 taught by Pietro Bianchi, Florida International University from Mar 2024 to Apr 2024.

117-038 Signet Jewelers: Assessing Customer Financing Risk

4

were the main driver, while earning power drove the trend for the 55-64 age group. The 25-34 age group spent an average of $786 on jewelry per household.33

Consumers seeking to buy jewelry had several choices, from specialty retail jewelers like Signet and Tiffany & Co. (Tiffany), large all-purpose retailers such as Wal-Mart and Costco, department stores such as Macy’s and J.C. Penney, online from Blue Nile, and TV retailers like QVC. Specialty jewelry retailers made up 39% of sales. Online retailing was growing, accounting for a 16% of sales by 2016, facilitated by improved technology and stronger consumer confidence in ecommerce.34

The retail jewelry industry was highly fragmented and competitive, with around 21,000 retailers at the end of 2014.35 Mergers and bankruptcies were common occurrences.36 Sector specialists reported that over 200 jewelry retailers in the U.S. closed their operations in the third quarter of 2015 alone. Specialty jewelers such as Signet, Blue Nile, and Tiffany competed on brand reputation, customer service and product innovation, with competition for engagement jewelry being especially intense.37

Blue Nile

In 1999, Blue Nile became the first company to sell diamonds online and soon became the world largest online retailer of diamonds and fine jewelry.38 In 2015, the company achieved net sales of $480 million and employed over 350 people. The company’s goal was to “be nothing less than the preeminent destination for diamond engagement rings and fine jewelry both online and off.” It advertised competitive prices and maintained low inventory since diamonds were purchased on a “just in time” basis following customers’ orders. Blue Nile aimed to maintain lower overhead than traditional jewelers and pass the savings on to the customer.39 Blue Nile was an online-only retailer until 2015, when it began testing a display case of bridal jewelry in two Nordstrom stores.40 Shortly after, Blue Nile opened its first display-only stores, called “Webrooms,” soon expanding to four different U.S. states.41 In the “Webrooms” customers could see and touch precious stones, and receive guidance from Blue Nile consultants. However, all transactions were still completed online.

Blue Nile did not extend credit to customers, but issued a private label credit card through a sponsoring bank. The card did not have an annual fee and awarded special access to cardholder-only offers and promotions. On Blue Nile’s balance sheet, trade accounts receivable were composed primarily of amounts due from financial institutions related to credit card sales.

Tiffany & Co.

Tiffany, founded in 1837, was a global specialty retailer headquartered in New York City. The company was the clear leader of a fragmented luxury jewelry sector, capturing a 14.2% market share compared to 5.8% for the next-closest competitor, Cartier.42 Tiffany focused on maintaining its leading position in the luxury market, projecting a brand image associated with “high-quality gemstone jewelry, particularly diamond jewelry; sophisticated style and romance; excellent customer service; an elegant store and online environment; upscale store locations; “classic” product positioning; and distinctive and high-quality packaging materials (most significantly, the TIFFANY & CO. blue box).”43 Tiffany maintained a flagship store on Fifth Avenue in New York City, which served as an attraction for international tourists and serious buyers alike.

Tiffany managed four product segments: fashion jewelry, engagement jewelry and wedding bands, statement jewelry, and non-jewelry accessories such as leather goods, timepieces, china, crystal, and fragrances.44 The company operated 124 Tiffany & Co. stores in the Americas, 81 stores in Asia-Pacific, 56 stores in Japan, 41 stores in Europe, and five stores in the United Arab Emirates.45 The company also sold merchandise through an ecommerce website.

For the exclusive use of J. Vega, 2024.

This document is authorized for use only by Jose Vega in ACG 6175 – Spring 2024 taught by Pietro Bianchi, Florida International University from Mar 2024 to Apr 2024.

Signet Jewelers: Assessing Customer Financing Risk 117-038

5

Tiffany offered a “Tiffany Select Financing” program on engagement rings and watches starting at $1,000 and all other purchases starting at $2,500.46 The receivables associated with Tiffany’s credit program amounted to $71.9 million and $75.2 million in fiscal years 2016 and 2015, respectively. Tiffany reported that 97% of those receivables were current, and its allowance for estimated losses was $1.1 million on January 31, 2017 (see Exhibits 7a-7c for Blue Nile and Tiffany financials).

Luxury diamond jewelry was Tiffany’s main revenue source, with input purchased globally and manufactured mainly in America.47 In 2015, engagement jewelry and wedding bands made up 29% of Tiffany’s global sales, with an average price of $3,300.48 Tiffany’s had an impressive legacy of jewelry design, dating back to the 19th century. The company’s website touted its legacy of design leadership, claiming to have in 1886 “introduced the engagement ring as we know it today. Previously, diamond rings were set in bezels. But Mr. Tiffany’s ring was designed to highlight brilliant-cut diamonds by lifting the stone off the band into the light.”49 In addition, throughout the company’s history, various organizations had commissioned Tiffany’s to create custom designs, such as the Lombardi Trophy for the NFL and the Great Seal of the United States, which appears on the one-dollar bill.50

Cohodes Takes Aim at Signet

Cohodes’ Thesis

Marc Cohodes began his finance career in 1982 at Northern Trust in Chicago after completing his undergraduate education at Babson College. In 1985, he joined David Rocker (MBA ‘69) at his new fund, Rocker Partners, and helped it grow from $20 million in assets to $700 million by 2000.51 By then, Cohodes had risen to general partner, specializing in short-selling.a He was described as a man “who knew every trick company executives used to make their operations look better than they actually were. [Cohodes] prides himself on being able to spot trouble.”52 In 2003, Cohodes gained national attention when he engaged in a long-term short-selling campaign against NovaStar Financial, a subprime lender that eventually sought Chapter 11 bankruptcy protection.53

Cohodes outlined his reasoning for short-selling Signet. First, according to Cohodes,

They [Signet] don’t really sell jewelry. You can’t appraise this stuff. If you go to Blue Nile, Tiffany, or the HBS jewelry store, and they sell you a ring with a 2-carat diamond with G-color, VS1-clarity, you can get it appraised and insured. You are buying an asset. It’s real. You may overpay, but it’s worth something. You can appraise it, you can sell it. You buy a Zales, Kay, Jared’s piece of jewelry, it’s not worth anything. It’s not real. I mean, it’s real, but it’s so marked up, there’s no jewel or gem value to what you’re buying.54

Cohodes believed that Signet was boosting sales by offering customers credit because of the low quality of the jewelry. He believed that 60-65% of the company’s business came from lending to customers and the sale of warranty-like Extended Service Plans (ESPs).55 Cohodes said:

A while ago, this guy [an activist investor] showed up and encouraged Signet to use credit to boost sales. Once you use credit to boost sales, it’s like being on drugs. It’s very

a Short-selling was a bet placed on a stock that, according to the short-seller’s research and belief, was overvalued. To short-sell a stock, investors borrowed a stock from someone who owned it, then sold it. If the short seller made the right prediction, the stock fell in value and the short seller bought the stock on the open market to return what they had borrowed. If the repurchase price was lower because the stock dropped, the short seller made a profit. If the price rose, the short-seller would need more money to “cover” the same number of shares and the short-seller would lose money. Jeffrey B. Little and Lucien Rhodes, Understanding Wall Street, New York: McGraw-Hill, 2004, p. 94.

For the exclusive use of J. Vega, 2024.

This document is authorized for use only by Jose Vega in ACG 6175 – Spring 2024 taught by Pietro Bianchi, Florida International University from Mar 2024 to Apr 2024.

117-038 Signet Jewelers: Assessing Customer Financing Risk

6

hard to get off. And that’s where they are. Even if they try to sell their credit book: Who wants it? And how much of a haircut will they demand in order to buy it? And it’s kind of like, ok, is the credit book worth 80 cents on the dollar, 60 cents on the dollar? Whatever it is, it’s not worth 100. At the end of the day the credit is no good. I come for you and say you owe me money, I want the ring back, and the guy says go fish, we broke up and she took the ring. You can repo a car that has title. You can repo a house that has title. How do you have credit or anything against some ring, necklace, earrings, or whatever?56

The media picked up on the concerns that Cohodes and other investors had with Signet’s business model. On February 15, 2016, Bloomberg News ran an article titled “Is Signet a diamond empire or finance company?” The article detailed how an aggressive credit policy had helped Signet become one of the largest jewelry companies in the U.S. but cautioned, “behind its sparkly empire lie consumer loans that bankers might consider subprime debt.”57 Analysts estimated that 35% of Signet’s receivables were owed by subprime borrowers. They estimated that if Signet was to sell its credit book it would have to take a 25% to 30% discount on its subprime receivables, while the discount on prime receivables would be no greater than 5%.58

Use of Recency Accounting

Cohodes was particularly concerned with Signet’s accounting of its receivables portfolio through the “recency” method.59 With recency accounting, the customer might only need to pay a portion of the amount due in a period for the account to remain current. As a result, the amount of delinquent loans was understated relative to the more widely used approach, the contractual method.60 For Cohodes, recency accounting masked the true condition of Signet’s credit portfolio:

No retailer uses recency accounting anymore. No one uses it. The question is, why do they use it and what would it look like if they account for it like other companies? And the credit trends are not going their way and that’s in a boom economy. So frankly, I don’t know how they can get out of this thing. I think they’re heading for a cliff. The cliff is, ‘we can’t sell the credit book, we can’t really extend more credit and our business is very credit dependent.’ It’s kind of like they’re going to have to go off of heroin cold turkey which is going to affect their sales and earnings. Then the stock falls and debt becomes an issue.61

One analyst noted that, “the percentage of non-performing loans as a percentage of Signet’s receivable balance has remained constant: at 3.8% in 2015; 3.7% in 2014; 3.6% in 2013; 3.7% in 2012” and that it was rare to see such limited movement in non-performing loans, suggesting that the trend may indicate the use of the recency method to produce stable numbers.62

In June 2016, Grant’s Interest Rate Observer reported on Signet, based on Cohodes research. The Grant’s report announced that “in-store credit facilitated no fewer than 61.7% of sales in the quarter ended April 30 [2016]” increasing from 52.6% in fiscal year 2007. The report also noted Cohodes research on bankruptcy filings that named Signet as a creditor. Through March of 2016, 3,274 individuals submitting for personal bankruptcy named Signet as a creditor, up from 2,663 in the fourth quarter of 2015 and 1,903 in the first quarter of 2015.63 Following the report, Signet stock fell 6.58%, to close at $92.23 on June 2, 2016.64

Extended Service Plans

There were additional concerns regarding Signet’s sales of extended service plans (ESPs). The ESPs functioned like a warranty and allowed consumers to access special services, such as the possibility of having a ring resized for life. ESP sales had grown in significance as a percentage of total sales from

For the exclusive use of J. Vega, 2024.

This document is authorized for use only by Jose Vega in ACG 6175 – Spring 2024 taught by Pietro Bianchi, Florida International University from Mar 2024 to Apr 2024.

Signet Jewelers: Assessing Customer Financing Risk 117-038

7

2014 to 2016.65 Most rings in Signet’s stores were sized 7 for women and 10 for men, requiring resizing upfront for most customers. Signet sales representatives would pitch the ESP to these customers since it was only slightly more expensive than the resizing alone. Grant’s Interest Rate Observer noted that ESPs were responsible for $348 million in revenue, equal to 3.5% of total revenue, in the fiscal year ending January 30, 2016.66 Signet did not disclose the profit margin on these plans, but Grant’s believed that they were a material portion of operating earnings.67

Cohodes was more colorful in his characterization of the ESP program: “Why you need an extended warranty on jewelry is beyond me. That’s the dumbest thing I’ve ever seen. Oh, it includes ‘free ring sizing.’ Come on now.”68

Two Recent Scandals

Compounding the criticisms of Cohodes and other short-sellers was a run of bad news for Signet. In February 2016, an arbiter ruled that a group of current and former employees could pursue their claims that Signet had discriminated against women under the Equal Pay Act. Several female workers had filed a lawsuit in 2008 alleging that Signet had paid them less and promoted them less often than their male counterparts.69 The arbiter’s decision came two years after other female employees accused CEO Mark Light and other top executives of bias and harassment.70 In February 2017, the Washington Post reported that “roughly 250 women and men who worked at Sterling allege that female employees at the company throughout the 1990s and 2000s were routinely sexually harassed.”71 This news led to a one-day stock price decline of 12.75%.72 Cohodes reacted to the scandal stating: “I’m not surprised. When I invest, I always look for bad management because bad people make for great shorts. They never disappoint. They not only mess up once, they do it over and over again.”

In May 2016, BuzzFeed, an internet media outlet, identified a group of Kay Jewelers customers that complained about receiving defective jewels after sending them in for repairs.73 More than 300 customers lodged complaints against Signet with the Consumer Financial Protection Bureau, lamenting that the diamonds did not sparkle as much, had imperfections, and even had different serial numbers.74 Analysts reported that customers were taking to Kay’s Facebook page to register complaints, focusing on “increased dissatisfaction among shoppers beginning in April […] reaching an average of 5.5 complaints per day.”75 Most complaints were about replaced gems, but branched out to other products, services, lost items, and bad repairs.76

Signet Responds

Recency Accounting

In March of 2016, Signet began to respond to concerns about its credit portfolio and recency accounting. CFO Michele Santana responded to criticisms, saying that “we [Signet] have provided and operated in-house credit for 30 years and it gives us a number of competitive advantages,” especially when it came to bridal sales. She continued: “due to our scale, we are able to administer our credit program very efficiently and effectively,” and that it was “designed for rapid repayment that minimizes risk and enables the customer to make additional jewelry purchases using their credit facility.”77

During the 2016 first quarter conference call in May 2016, Santana responded to questions about the use of recency accounting:

For the exclusive use of J. Vega, 2024.

This document is authorized for use only by Jose Vega in ACG 6175 – Spring 2024 taught by Pietro Bianchi, Florida International University from Mar 2024 to Apr 2024.

117-038 Signet Jewelers: Assessing Customer Financing Risk

8

At the end of the day, regardless of recency or contractual, whatever method you’re on, the financial results are going to yield the same answer. The provision will be the same, our bad debt expense will be the same, but to that point, the reason why we use our recency is, one, we have done it since the beginning of time and it really has worked well for us over the years with the type of lending that we do. Jewelry lending is that emotional connection and it does optimize our collections for us… It gives the customer some flexibility based on the disciplined criteria that we had outlined of our recency ageing, what a customer has to remit to stay current. It leaves that customer in good standing if they are having maybe a challenging month where they can’t remit a full payment, and that psychology and that flexibility of working with that customer puts that cus

Related Tags