Conn’s (CONN): A Retailer Targeting the Subprime Segment
“Our confidence is shattered,” said Laura Champine, an analyst with Canaccord Genuity Corp., with regard to Conn’s.
Such quotes get our pulse racing. When Wall Street gives up and throws in the towel, we make money.
What prompted such a dramatic pronouncement? Quarter 3 of fiscal year 2015 was nothing short of a disaster for Conn’s due to problems in its loan portfolio. Its credit segment operating income dropped by $43.6 million, driven by an 82% increase in the provision for bad debts over the prior quarter. Further, delinquency rates were up 130 basis points to 10%. The company admitted that losses were being realized quicker than anticipated and they did not believe there will be any improvement over the next 12 months.
The key question going forward, therefore, is this a temporary problem that can be fixed or does it represent a permanent reduction in returns?
CONN is a specialty retailer of appliances, furniture, mattresses and electronics with 89 locations in Texas and the Southwest. Significantly, CONN finances 77% of customer purchases through its proprietary subprime credit portfolio.
The retail industry has been transformed over the last 20 years. Amazon.com and other players have emerged while former stalwarts like Circuit City, K-Mart, Borders and Sears are dead or dying. Consumers and the wider economy have reaped the benefits of this transformation. An overview of the retail industry over the last 20 years is beyond the scope of this blog post. Suffice it to say that the industry is very crowded and ruthlessly competitive.
Conn’s competes against Sears, Wal-Mart, Target, Sam’s Club and Costco, specialized national retailers such as Best Buy, Rooms To Go and Mattress Firm, home improvement stores such as Lowe’s and Home Depot, and locally-owned regional or independent retail specialty. Conn’s also competes against companies offering credit constrained consumers products for the home under weekly or monthly rent-to-own payment options. Such companies include Aaron’s and Rent-A-Center, as well as many smaller, independent companies.
How’s does Conn’s survive in this jungle? It has carved out a unique niche for itself. Conn’s has a narrowly tailored strategy to serve the financially strapped working class household. These customers represent a unique segment of the middle-income market that is underserved by traditional banking institutions. The core consumer base is comprised of working individuals who typically earn between $25,000 and $60,000 in annual income. The typical customer is unbanked and has credit scores between 550 and 650. (See Figure 1.) This segment of the population has limited disposable income after rent, transportation, food, health care and communications. In and of itself, this is not sufficient to set Conn’s apart from the competition. The unique feature of Conn’s strategy is the subprime financing arm. By offering in-house financing, Conn’s makes an aspirational lifestyle available where it would otherwise not be. Their competitors typically do not provide a similar credit offering.
This segment of the business is worth a closer look.
The Subprime Financing Segment
Conn’s provide in-house financing to individual consumers on a short-term basis (maximum initial contractual term is 32 months) for the purchase of durable products for the home. A significant portion of the credit portfolio is due from customers that are considered higher-risk, subprime borrowers. The financing is executed using an installment contract, which requires a fixed monthly payment over a fixed term. Conn’s maintains a secured interest in the product financed. If a payment is delayed, missed or paid only in part, the account becomes delinquent, in-house collection personnel attempt to contact the customer once their account becomes delinquent.
Conn’s generally provide for interest at the maximum rate allowed by the respective state regulations, except for Arizona and New Mexico where they charge 26% and state regulations do not generally limit the interest rate charged. The average yield is 17-18%.
Conn’s also promotes the sale of many of products through advertised short-term, no-interest programs. The majority of these accounts have a 12-month term, with the balance carrying a term of six or three months. Minimum monthly customer payments are required. If the customer is delinquent in making a scheduled monthly payment or does not repay the principal in full by the end of the stated term (grace periods are provided), the account no longer qualifies for the no interest provision and the account reverts back to the terms of the executed retail installment loan contract. A significant number of customers who purchase through promotional programs meet the terms of the program, resulting in the payoff of those accounts and thus an increase in the velocity of portfolio turnover.
The specialty finance industry serves the financial needs of the unbanked or underbanked. The unbanked are those households with neither a saving nor a checking account at a financial institution. The under-banked may have an account at a bank but choose to use some non-bank services including money orders check cashing services, remittances, payday loans, rent-to-own services, pawn shops, or refund anticipation loans (collectively the “specialty finance industry”). These consumers conduct some or all of their ﬁnancial transactions outside of the mainstream banking system.
The specialty finance industry encompasses a wide variety of products, services and channels. Some specialty finance companies are engaged purely in providing alternative financial services through retail storefronts. Others serve customers via the Internet. Some target both retail and Internet channels. Some focus only on regional US markets while others are in one or more international markets. (See Figure 2.)
Installment lending to non-prime consumers is one of the most highly fragmented sectors of the consumer finance industry. Installment loans, the product Conn’s offers, are estimated to be provided through approximately 5,000 individually-licensed finance company branches in the United States.
The trend of consumers to opt out of traditional banking has accelerated since the financial crisis as the big banks turned away from non-prime segment. “Certainly conventional lenders over the past five to six years have become increasingly reluctant to advance credit on all fronts,” one pawnbroker said. “The time it takes those institutions to make decisions has lengthened as their appetite has shrunk. You can walk into our location in New York and walk out with your funds in an hour or two.”
Subprime lending fills an important need by allowing borrowers with tarnished credit or without access to traditional financial institutions to purchase goods.
Total Addressable Market
Conn’s totally addressable market represents approximately one-third of the population of the US.
According to a 2013 survey by the Federal Deposit Insurance Corporation, the unbanked make up 7.7% of the US population or about 10 million households and the under-banked are another 20%.
The financially underserved market generated an estimated $103 billion in 2013 in fee and interest revenue. Revenue grew by 7.1% from 2012 to 2013. Market revenue has experienced significant and steady growth, expanding 26% from 2009 to 2013, a compounded annual growth rate (“CAGR”) of 6%. (See Figure 3.)
Macroeconomic factors seem to be driving this growth, particularly growing income inequality in the US. The causes and persistence of this trend is far outside my circle of competence but the basic facts are fairly clear. (See Figure 4.)
While demand has grown, the number of lenders supplying credit to this sector of the population has decreased. The competitive landscape has changed dramatically since 2008. The industry’s traditional lenders, including Wells Fargo, HSBC, Citi and AIG, have recently undergone fundamental changes, forcing many to retrench and in some cases to exit the market altogether. Tightened credit requirements imposed by banks, credit card companies, and other traditional lenders that began during the recession of 2008-2009 have further reduced the supply of consumer credit for non-prime borrowers. In addition, recent regulatory developments create a dis-incentive for these lenders to resume or support these lending activities.
Concurrently, subprime credit card providers have exited the industry. The Credit Card Accountability Responsibility and Disclosure Act of 2009 added new restrictions on late fees, interest-rate increases and other pricing tools used by card issuers before the financial crisis.
Additionally, states have cracked down on payday lenders, effectively pushing payday lenders out of many states.
As a result of the reduced output of these companies, access to credit has fallen substantially for the non-prime segment of customers. The alternatives are not particularly attractive. Many subprime borrowers have turned to title lenders and subprime auto lenders, both of which are experiencing booms. New entrants are also entering the market.
This large and growing number of potential customers in Conn’s target market, combined with the decline in available consumer credit, provides an attractive market opportunity for Conn’s business model.
What Are The Benefits Of The Subprime Program For Conn’s?
I think I have made the case that there is both a demand for subprime lending and a shortage of supply. It would seem, therefore, that Conn’s is well-positioned to take advantage of this imbalance. How, specifically, is the credit segment helping rather than hurting Conn’s?
First and foremost, the subprime program is a key aspect of Conn’s strategy of focusing on the working class segment of the population. It is a key differentiator and creates a narrow moat around their business. Where do we see this in the income statement?
- Increased Sales Growth – Conn’s is growing much faster than its competitors and it attributes the growth to the availability of credit. Below average income customers can purchase branded products with the help of Conn’s financing.
- Higher Prices – Conn’s financing program allows customers to buy higher quality products than they could otherwise afford resulting in higher prices across the board. For example, their average television selling price is $1,179 versus $465 for the market as a whole. This phenomenon drives in Conn’s sky high gross margins – 40% in FY2014 up from 26.5% in FY 2011 prior to the CEO change.
Figure 5, Conn’s Gross Margin vs. Competitors
- More foot traffic – The credit program also drives more foot traffic into the stores. Fully 56.6% of Conn’s borrowers pay in-store. This leads to more impulse buys while the customers are in the store creates a relationship with Conn’s.
- Repeat Business — The credit program also leads to a high level of repeat business. After the initial purchase, the average customer buys 2 more times with 5 years. For those that make a second purchase, those customers purchase 4 times within 5 years. Rate of repeat business is even higher in mature markets.
- Credit Insurance – The credit program also drives high margin sales of credit insurance. Part of Conn’s strategy to control delinquency and the net charge off rate is sale of insurance. Conn’s requires customers to provide proof of property insurance coverage to offset potential losses relating to theft or damage of the product financed. Conn’s offers basic payment protection insurance products. Conn’s receives sales commissions from the third party insurance company. In fiscal year 2013, approximately 84.8% of Conn’s credit customers purchased credit insurance. Commission revenues from the sale of credit insurance contracts represented approximately 3.7%, 2.9% and 2.5% of total revenues for fiscal years 2014, 2013 and 2012, respectively.
Deterioration of the Credit Portfolio
The facts above notwithstanding, the quality and performance of Conn’s credit portfolio has noticeably deteriorated over the past 2 to 3 years. This drop in quality coincides with the arrival of an aggressive new CEO in the first quarter of fiscal year 2012. Since his arrival, the size of the loan portfolio has exploded, going from $625 million to $1.2 billion – a CAGR of 5%. Further the percentage of customers choosing to finance their purchase jumped from 64% to 77%. Along with the increase in size there has been a concomitant deterioration in quality:
- The net charge off rate increased from 6.8% in Q1 2012 to 8.9% in Q3 2015
- The average down payment dropped from 8% to 3.6% in Q3 2015
- The bad debt provision as a percentage of the loan portfolio jumped from 1% to 6% in Q3 2015.
- The percentage of balances 60+ days past due rose from 7% at the end of 2005 to 10% in Q3 2015.
- Over the last 5 years, the average credit score of loans originated has bounced around a lot but dropped a great deal before rebounding lately.
- After dropping, the percent of loans being modified and extended (“re-aged” in Conn’s parlance) has begun increasing again.
What’s Going On Here?
These statistics can certainly give rise to an inference that Conn’s has intentionally loosened underwriting requirements in an attempt to boost sales. The arrival of aggressive new management in 2012 and the subsequent deterioration of the loan portfolio only adds to this concern. This view is further corroborated by the assertions of former employees in a recent class action suit.
Compounding the problem, management refused to admit there was an issue and seemed to take steps to cover up the deterioration. The most questionable of these methods is “re-aging.” Conn’s regularly extends or “re-ages” a portion of delinquent customer accounts. Generally, extensions are granted to customers who have experienced a financial difficulty (such as the temporary loss of employment), which was subsequently resolved and the customer indicates a willingness and ability to resume making monthly payments. These re-ages involve modifying the payment terms to defer a portion of the cash payments currently required of the debtor to help the debtor improve his or her financial condition and eventually be able to pay Conn’s. This practice is derisively known as “extend and pretend.” Such loan extensions can be used to temporarily cover up bad loans.
Other methods include:
- Under providing for loan losses. After dropping by 42% in Q1 2015, the loan provision increased by 78% in Q2 and 82% in Q3. This would seem to suggest the company was hiding the poor quality of the loan portfolio and was forced to dramatically increase the provision when the numbers left them with no choice.
- Delay charge-offs. Again this puts off the day of reckoning.
- Focus on sales growth rather than loan performance to distract analysts.
- Blame other factors including poor software implementation, an under-staffed collections department or cold weather and higher heating bills. As recently as March 27, 2014, management was still saying that “the unexpected delinquency increase” was not a result of deterioration in the underlying credit quality or a change in underwriting standards.
Now, the Company, however, acknowledges that the credit quality has deteriorated but places the blame on the higher proportion of new customers in the loan portfolio and the product mix. New customers default on loans at twice the rate of existing customers. The proportion of new customers in the portfolio will drop going forward as growth slows. The product purchased also seems to impact the default rate. Conn’s experiences a much lower rate of default on furniture and appliance sales than on consumer electronics and home office. Conn’s is putting much more emphasis on furniture sales as a result. Thus, the problems should resolve themselves in the medium term.
Does It Matter?
There is no doubt that greatest risks to Conn’s are poor underwriting, insufficient loan loss reserves and excessive debt. Companies in the subprime area often fail due to excessive risk taking. There is no need to review that gory history here.
However, unless you think management are crooks that looted the company and will now walk away from it, does either view really matter for assessing the value of the company in the future? If they did intentionally lower credit standards, that mistake has caught up with them. Current management has suffered a 75% drop in the stock price and the CFO paid with his job. They have taken a number of steps to correct the problem.
Further, unless there is more bad news to come, the current losses can be borne without a problem by the company. The company estimates that the credit segment will breakeven with a charge off rate of 11% and the company itself will stay profitable unless the charge off rate climbs to 28%. (Remember: Even the bad 2014 and 2015 vintage loans are estimated to have just a 10% net charge-off rate.)
All in all, it seems like a classic pig in the python situation. That is, the company will have to bite the bullet and deal with a sharp increase in charge offs in the short term, but the situation will resolve itself thereafter. Further, one important distinction between Conn’s and other subprime lenders is the short term duration of their loans. The weighted average term of a loan at origination is 24.5 months with a maximum of 32 months. The average account remains outstanding for only 11 months. These are not 30 year mortgages. Any bad underwriting shows up quickly (as we have seen). Old loans will not drag down the performance of the company for years to come.
Can Subprime Financing Form the Basis of a Moat?
Nothing is more commodity-like than a loan. Can a lending business really be the basis of a moat?
A lender can only have a moat if it maintains a conservative and disciplined approach to credit risk. As the inimitable Charlie Munger put it: “A friend of mine won’t touch banks. His attitude is that sooner or later the bastards will go crazy. I think that’s irrational. You have to be able to recognize the ones that stick out. Wells Fargo and US Bancorp avoid stupidity better than most. And Wells admits that it had its head up its ass when it made some of its mortgage loans. They know it wasn’t their finest moment. I’m comfortable with people like that.”
The best lenders have a high quality underwriting supported by proprietary analytics and a “high touch” approach. They do not rely exclusively on FICO scores. The employees are long tenured and have substantial local market knowledge and experience. The best build a consistent and deeply-engrained underwriting culture that produces consistently well-underwritten loans. This qualitative knowledge is complemented by proprietary technology, data analytics and decisioning tools.
As I have written before, a conservative and disciplined approach to credit risk is, in and of itself, a durable competitive advantage.
In essence, a sound subprime lending operation requires four disciplines: (1) An understanding of all exposures that might cause a borrower to default; (2) A conservative evaluation of the likelihood of any exposure actually causing a default and the probable cost if it does; (3) The setting of an interest rate that will deliver a profit, on average, after both prospective loss costs and operating expenses are covered; and (4) The willingness to walk away if the appropriate interest rate can’t be obtained.
Conn’s seems to have failed most egregiously at #1 and #2. That is, in the rush to increase sales, the company overlooked problems that could signal a high likelihood of default. They also failed to properly provide for loan losses.
While I am confident that Conn’s see the error of their ways and is taking steps to tighten up the underwriting process, I remain concerned about their centralized approach to assessing credit risk. The best lenders typically live in the communities they serve, value the face-to-face interactions with customers and build long-term relationships. This face-to-face interaction significantly enhances their ability to assess customers’ household budgets and ability to repay their loans. The knowledge gained and relationships built during the face-to-face improves loan performance. Branches should not only originate but also service loans through early-stage delinquency. Once this localized model is in place, certain functions and activities can be centralized and shared to deliver cost efficiencies and risk and compliance controls. In my view, this is the only way to really drive low default and delinquency rates. I think Conn’s will continue to have problems until and unless more underwriting is moved to the stores.
A second issue is whether this approach could be easily copied by Conn’s myriad competitors. In other words, couldn’t Walmart easily add a credit operation and put Conn’s out of business?
- Retail/Lending Incompatibility – Is it possible that Walmart and other companies who certainly have the resources to start in-house financing operations have shied away from it because lending and sales are fundamentally incompatible? Human nature being what it is, is it simply unavoidable that lending decisions will be compromised by the pressure to close sales? Conn’s claims that lending decisions are made entirely independent of sales personnel. But, is this believable? The class action suit referred to above certainly suggests otherwise. I don’t know the answer to these questions, but it is certainly a cause for concern.
- Access to Capital – Another barrier to entry is access to capital. This certainly doesn’t affect Walmart, Target or Amazon, but it might keep smaller competitors out of the market.
- Experience/Learning Curve – Retailers credit experience. Underwriting, collecting and managing a credit portfolio require significant experience and trained portfolio. Conn’s has 50 years of experience in this area and still has problems.
- Threat to brand – Some competitors might see a focus on this customer segment and subprime lending as a threat to their brands image. Because they are trying to appeal to multiple segments, the negative connotations of subprime might tarnish certain segment’s view of their company and ultimately do more arm then good.
- Real Estate Strategy – Some competitors real estate strategy might run counter to an appeal to subprime customers. Stores in upscale markets simply can’t draw this segment in sufficient numbers to make it worthwhile.
Local Economies of Scale/Store Density
A second key component of Conn’s strategy is economies of scale driven by high levels of store density. Conn’s typically enters a market with a scaled presence to efficiently leverage advertising spending, regional management and delivery and distribution infrastructure. Within a short time period, usually not exceeding 18 months, Conn’s grows the store count in the new market to optimize the economies of scale.
To be sure, the economies of scale benefit is not yet showing up in the income statement. Specifically, Conn’s Sales, General & Administrative expense (“SG&A”) as a percentage of revenue is much higher than its competitors. (See Figure 7.) I see this mismanagement as creating an opportunity for margin expansion.
Figure 7, SG&A Expense as % of Revenue
Conn’s new store economics are too good to be believed. Conn’s claims that they can invest approximately $1 million in a new store and make back the investment in 3-6 months. How is this possible? I’m pretty sure even I could raise the money on Kickstarter to open a new store with this return! One issue, of course, is that the “cash payback” indicated in the slide is not really true. All of this cash would have to be reinvested in inventory and the loan portfolio. (See Figure 8.)
Beyond that, though, competitors report similar new store economics. For example, HH Gregg states: “Our stores typically generate positive cash flow within three months of opening and provide a cash payback in less than three years.”
Aaron’s and Rent-a-Center show present similar data. (See Figures 9 and 10.)
Figures 9 & 10
Unless all of these managements are lying, the economics must be this compelling.
Management believes that the US market can ultimately support 400 stores.
Balance Sheet Risk
The doomsday scenario facing potential or current Conn’s investors is that the loan portfolio continues to deteriorate and the company finds itself over-extended and over-leveraged. Specifically:
- Credit quality continues its dramatic decline. This would seem to only be possible if management has still not revealed the full extent of problems in the loan portfolio or if they are committing fraud.
- Creditors are no longer willing to lend to Conn’s.
- Conn’s is forced to reduce its lending.
- Sales are reduced as a result
- A death spiral results.
Yet, a comparison with some of its peers does not find Conn’s more in debt or holding a far riskier loan portfolio. (See Figure 11.)
Figure 11, Comparison of Key Balance Sheet Risk Indicators
Thus, we are faced with a couple of starkly different views of the company. Management would have us believe that the higher loan losses are an unavoidable byproduct of rapid growth and in no way do the losses affect the survivability of the company in the short or long term. Bears would have us believe that the company is merely a thinly disguised predatory lender taking advantage of vulnerable borrowers. They rushed to get as many iffy borrowers into loans as quickly as possible. They then hid the instability of the loans using a number of accounting tricks. Using these tricks, they could then claim that they had more well performing loans than they really did. This tricked shareholders into believing that the company’s loan portfolio was stronger than it really was.
While the bear view is not unreasonable, my main hypothesis is that charge-offs will rise in the short-term but not nearly enough to drive the company into a loss, trip debt covenants or even foretell a prolonged period of depressed profitability. As stated above, Conn’s loans are short-term. They have reacted quickly to adjust underwriting requirements. Their basic business model and strategy is still strong and fundamentals are unlikely deteriorate dramatically. Again, the charge-off rate would have to nearly triple just to drive the company to breakeven. This seems very unlikely.
I made a few simple assumptions to project out Conn’s earnings in 2025.
- Same Store Sales = 10% per year (10 year average = 5%, 3 year average = 14.5%). This is conservative as management believes. Tighter underwriting creates a 5-7% headwind.
- New Stores = 15/year
- New Store Revenue = $15 million per store in 2016 and increases by 10% per year
- Gross Profit % = 40% (10 year average = 39%, 3 year average = 46%
- SGA % = 27% (10 year and 3 year average = 29%). I assume 200 bps of expense reduction due to increase economies of scale.
- Loan Portfolio Growth = 10% per year (10 year CAGR = 11%, 3 year CAGR = 29%)
- Bad Debt Provision = 8% of Loan Portfolio (10 year average = 5%, 3 year average = 9%). This management’s stated long-term goal.
- Tax Rate = 35%
These assumptions yield the following 2024 earnings calculation:
At a P/E of 12, the value of Conn’s would be $1.5 billion – more than double today’s value.
Note that I have not assumed any shareholder return in the form of dividends or share buy-back. Management has stated that they can get to a 20% ROE but I have assume that 100% of earnings are invested in (1) customer loans, (2) new store development, (3) maintenance capex and (4) inventory. I have also not assumed any increase in the share count.
There are two narratives at work with regard to CONN. The first is that CONN is a slow-growth big box retailer with a sleazy subprime financing arm with loose underwriting requirements that artificially pumps up sales. This collapse in the loan portfolio was simply inevitable. The second view is that CONN is and remains a fine company and the market greatly over-reacts to any bad news related to subprime. Whether it is the high interest rates, the sales pitches than can seem déclassé, or the accusations of “predatory” lending, this industry cannot seem to stay out of the news.
Given the bad experience with subprime mortgages, an unexpected rise in Conn’s charge-offs is sufficient to spook Mr. Market. As outlined above, I think this is an oversimplification and a poor analogy. In summary, I like Conn’s at this price for the following reasons:
- Combining sales of household durables with subprime lending is the basis for an interesting strategy.
- The total addressable market for this strategy is huge
- The number of lenders to this market has shrunk creating a fundamental mismatch between supply and demand.
- While this imbalance will certainly attract new entrants, there are a number of barriers to entry that reduce this threat.
- These factors combine to create a huge opportunity for Conn’s.
 Conn’s fiscal year ends on January 31.
 The weighted average credit score of accounts originated in Q3 2015, the most recent quarter, was 608.
 Springleaf Holdings, Inc., Form S-1, Page 2.
 NY Times, “Need Cash? Own a Bentley? Take a Pawn Ticket,” January 12, 2014.
 The NY Times has been running a not particularly flattering series on these lenders: http://dealbook.nytimes.com/category/series/driven-into-debt/
 Interesting interview with the found of one of the new entrants can be found here: http://www.vox.com/personal-finance/2015/1/22/7871367/raj-date-mortgage-payday
 Conn’s December Investor Presentation, page 8.
 While common in the industry, sales of credit insurance are not without controversy due to poor disclosure. Many state authorities are looking at them and litigation and/or regulation is possible.
 See here: http://graphics8.nytimes.com/packages/pdf/business/056SecondAmendedComplaint.pdf
 Conn’s December 2014 Investor Presentation, page 33.
 I got this concept from Bruce Berkowitz. Here is a summary of his view of Citigroup in 2010: “In his view, the financial services industry is like a python and the bad debt moving through the system is a pig that the python has swallowed whole. When the python first swallows the pig there is a huge lump in its body that prevents it from moving as freely as it normally would. At this point, Berkowitz believes that most of the bad debt has been identified and that it is in the process of being purged from the balance sheets of many financial institutions. In essence, the pig is now moving down the python, meaning that it will only be a matter of time before the snake is able to start moving freely again. As long as the pig continues to move down the python things should improve for financial firms. This dynamic was seen in the latest round of earnings from the sector, with both Citigroup and Bank of America reporting an easing of loan losses and delinquencies, and a dramatic improvement in earnings.”
 But remember that about 16% of all loans are modified and extended beyond this.
 David Einhorn, a significant Conn’s shareholder, summed up this view after a bad Q4 2014: “In February, the company announced 33% comparable store sales growth in Q4 with strong gross margins. However, it also announced increased credit losses and reduced earnings guidance from a range of $3.80-$4.00 to a range of $3.40-$3.70 for calendar 2014. Given the market’s past experience with deterioration in subprime credit, the stock reaction was severe: The price fell from $79 at the start of the year to $32 on the news. We believe that this is a retailer with 15-20% unit growth and current double digit comparable store sales growth, and that the market overreacted to moderately bad news. We acquired shares at an average price of $35.49 and they ended the quarter at $38.85.”