Description

Summary: Rumors of the FICO Score's demise are greatly exaggerated and shares look attractive. You are getting a chance to own a dominant franchise with a durable growth runway at an attractive valuation with the software business thrown in for free. 

What is happening? Since peaking in late July at ~$550, FICO shares are off 27% due we think to a couple of negative news items, a mixed fiscal 3rd quarter result, and the startingly strong performance of recent IPO Upstart:

(1) News that Synchrony transitioned away from FICO Score. Specifically, on their 2Q call, credit bureau Transunion disclosed that a Top 5 issuer had migrated away from the FICO score to rival score VantageScore (Synchrony Financial, the largest provider of private label credit cards in the US). TRU also noted that lenders are 'thinking hard about [transitioning to VantageScore] or in the process of migrating to Vantage.'

(2) A WSJ article titled 'FICO Score's Hold on the Credit Market is Slipping' was published. (https://www.wsj.com/articles/fico-scores-hold-on-the-credit-market-is-slipping-11627119003) The article highlighted the Synchrony news, anecdotal commentary from other very large lenders that are now relying less on FICO scores for originations and monitoring, and the potential that Fannie and Freddie will allow lenders to use different credit scores to underwrite mortgages. 

(3) Mixed FY3Q results with Scores results strong but offset by a lot of noise in its Software business results.

(4) Since its IPO in late 2020, Upstart (UPST) is up almost 16x. Upstart leverages AI on non-traditional underwriting variables to more accurately price risk for unsecured consumer loans than traditional methods. Notably on its 2Q call, UPST disclosed that one of its banking partners recently eliminated any minimum FICO requirement for its borrowers.

Our take on point (1): There is no love lost between FICO and TRU. There had been an ongoing dispute over royalties that involved lawsuits and (FICO suspects) TRU pushing for a DOJ investigation into FICO (this was closed last year with no issues). Ultimately, TRU was forced to pay owed royalties a year ago and it seems pretty clear they aren't happy about that. Re: Synchrony - FICO IR told me that this was the company that was in the middle of the royalty dispute and that they had been trying to switch off FICO for ~5 years. Importantly, this wasn't new news! Synchrony switched over ~1.5 years ago and it was disclosed in their 10K this year. Synchrony was <1% of revenues, and importantly, FICO sees no signs whatsoever of other large clients switching off the FICO score and thinks that it'd be very hard for a bank to undertake something like this without FICO knowing about it. 

Our take on point (2): The article contained a number of allegations that we would consider false, misleading, and/or old news. In order: 

• Capital One Financial Corp doesn’t use FICO scores for most consumer lender decisions – A review of their annual report suggests this is misleading and/or completely false. Capital One cites FICO 15 times in its annual report and flat out says that borrower credit scores are the key indicator it looks at when assessing borrower risk profiles. Note – it doesn’t specifically say FICO score but the table highlighting credit score distributions specifically cites FICO® trademark.

• Synchrony switched away from FICO – This is true but it happened 1.5 years ago and was in the works for ~5 years.  The timing of the WSJ citing this is interesting as it came RIGHT before TRU held their 2Q call and it probably allowed TRU to throw some dirt on FICO. There is clearly no love lost between the two companies.

• FICO is becoming a smaller factor in some underwriting decisions at JPM and BAC – This is probably true but who cares? It doesn’t impact the number of scores being pulled. FICO scores are referenced 14 times in Bank of America’s annual and 25 times in JPM’s. Banks have a lot more data they can use to make underwriting decisions and monitor credit accounts, so it is only natural that FICO ‘market share’ of the underwriting decision is lower today than it was a decade ago. This doesn’t mean that the FICO score isn’t still used in each of those situations, it just means banks are using additional metrics in addition to the score now. The cost of a bad customer is far higher than the cost of a FICO score, and with banks paying up to $200 for a credit card customer lead, the FICO score of just pennies is negligible.

• Fannie and Freddie are evaluating whether to allow lenders to use other scores when evaluating mortgage applicants – This is not new news and unlikely to have much financial impact on FICO, if any. The article also falsely makes the point that FICO’s dominance in securitization markets is dependent on Fannie and Freddie mandating that lenders use FICO scores if they want to package loans and sell to them. The real story is Fannie and Freddie mandated that lenders use FICO scores to package loans BECAUSE FICO was already so dominant in every other aspect of consumer lending.

• Bank credit executives say their own internal data and proprietary scores are more reliable than FICO scores – We file this under the ‘And your point is what?’ category. What credit executive is going to tell you that their job is useless and could be outsourced to a 3^rd party score? And as mentioned earlier, banks have more data on their own consumers so their own scores are probably more accurate – but it doesn’t help for customers outside the bank. It is highly likely that their models incorporate a FICO score anyways.

• FICO is becoming a smaller factor in underwriting decisions at Citizens Financial Group – WRONG. From their 10-K – “For retail loans, we utilize credit scores provided by FICO which are generally refreshed on a quarterly basis and the loan’s payment and delinquency status to monitor credit quality. Management believes FICO credit scores are considered the strongest indicator of credit losses over the contractual life of the loan as the scores are based on current and historical national industry-wide consumer level credit performance data, and assist management in predicting the borrower’s future payment performance."
  
  
• FICO’s newer scores that help banks underwrite loans for people without traditional scores isn’t gaining traction – Banks move VERY slow out of conservatism. The most popular FICO scores used by banks are FICO 2, 4, and 5 whereas FICO is on version 10 right now. It is not a surprise that banks aren’t in a rush to adopt the latest and greatest products and are choosing to look at only their own customers without traditional credit scores right now

Our take on (3): The software transition is easily the most frustrating aspect of the investment thesis, particularly as the company deploys 100% of its excess capital into share buybacks so the alternative to these investments would be a larger per-share ownership in the attractive scores business. With that said, the Scores business on its own supports the investment outlook at current levels, and at least management does appear to be done increasing the level of investment to support this transition. It noted at a recent investment conference that it did not expect Software margins to get worse, and the company recently de-emphasized lower margin professional services business and divested its non-core collections and recovery software business. So at worst, it does not appear to be an increasing drag on results and there is always optionality should the company successfully become the leading decisioning analytical platform and expand into tangental end markets besides financial services. 

Our take on (4): While yet to go through a full credit cycle, Upstart appears to have an interesting product and has seen the best traction offering underwriting personal loans for borrowers with low credit scores. It also happens to use the FICO score as one of the inputs for its AI model. On a recent conference call, FICO's CEO noted that many of the would-be fintech competitors also happen to be their customers, and that if the Score is available it would be nonsensical to not use the score given how low the cost is relative to the cost of a bad loan. 

So with that lead-in, we think the story is simple. The market is overreacting. This is frankly reasonable given how important the FICO Score is to the company's overall value, but therein lies the opportunity. This isn't the first time, either. It's hard to see on a long-term chart because the curve is so steep, but FICO seems to have a 20%+ drawdown every year over the past decade plus when it has been crushing the broader market. In each of those circumstances, these sell-offs proved to be great buying opportunities. We see a similar outcome right now. 

Brief Overview of Business / Investment Opportunity

Business Description: FICO is a leading information services company serving primarily the financial services industry. It is best known for its FICO credit score that is used in the majority of credit decisions in the US, but the company also has a large software business selling fraud, originations, customer management, and compliance solutions primarily to financial institutions. The common theme across both businesses is that FICO's solutions help decisions make faster, more accurate decisions.

Software results muddied by a license to subscription conversion... It's software business enjoys leading market positions within its core financial institution customers but segment results have been muddied by a multi-year transition from offering on-premise point solutions to a cloud-based platform that will enable the business to generate more business with smaller banking customers and expand into more verticals. This has been a long and costly endeavor, with Software (Applications + DMS segment) FY21E operating profits 51% below the company's FY14 high of $174mn, and margins have declined from the 25-30% range to an estimated 12% this fiscal year.

...but Scores has been carrying the weight: Offsetting this headwind, however, has been the exceptional performance of the Scores segment, which has grown sales & operating profits at a 20% and 22% CAGR, respectively, since FY14. The Scores segment now accounts for well over 100% of consolidated profits. 

On a consolidated basis, FICO Scores strength has more than offset elevated investment in the software business. Since FY14, consolidated pre-tax profits have increased 14% annually (through FY20) while adjusted EPS has increased 17% and free cash flow has increased 13%. More recently, earnings and cash flow have accelerated as investment spending for the software transition has plateaued at the same time FICO implemented special pricing adjustments in its Scores segment. Adjusted Earnings and FCF increased 30% and 34% annually from FY18-20 and are +41% and +57%, respectively in the first 9 months of FY21. 

Looking ahead, we see more of the same. Scores should continue to put up strong growth at near 100% incremental margins driven by consumer lending growth, special pricing (~$50mn annually) and its partnerships with Experian (ex. EXPN's credit monitoring program and Boost).

Software remains the wildcard as several years into this transition, it does not sound like the business is close to turning a corner. Fortunately, management does appear to be done increasing the level of investment, noting at a recent investment conference that it did not expect margins to get worse. And the company has recently de-emphasized lower margin professional services business and divested its non-core collections and recovery software business. 

Importantly, we do not need a recovery in software to justify the investment. On our estimate, pre-tax profits can increase mid-to-high teens next few years even with software profits flat. 

Valuation: At $404, shares are trading at <25x trailing FCF and ~22x consolidated EV to Scores EBIT. With the Software business essentially not even covering its share of corporate overhead, it reasonable to assume 100% of FCF is the Scores business, and 25x seems too low considering its market position and growth outlook. 

The Software business is harder to value. The company is in the midst of an ambitious plan to change from on-premise point solutions for financial institutions to a cloud-based decisioning platform for all types of B2C companies. It is currently finishing building out the platforms with the goal of turning its internal APIs - which it uses to build solutions for its enterprise customers - outward such that independent software vendors can build out solutions on their platform and take them to their customers directly. This is extremely ambitious and management has already walked back growth targets previously, so some skepticism is warranted. At a recent investor conference, management was asked about whether it's 5-yr plan was too slow given the accelerating pace of innovation and it sounds like investors think the business should be sold and/or they should partner to get where they need to be faster. We share investor frustrations but fortunately think the value of the Scores business supports the investment case on its own so we can be patient with the software business. If we had to value this piece of the business, a conservative (potentially punitive?) approach might be just to value the SaaS subscription component (30% of Software revenue) and ascribe no value to the legacy software business. As of FY20, the company had ~$235mn in SaaS subscription revenues (growing 11%). Around that growth rate, SaaS businesses tend to trade in the 4-10x sales, so call it $1bn to $2.5bn in value for this business ($35-90/share). 

Appendix: Why is Scores Such A Great Business?

• Dominant share: FICO has over 90% market share of credit decisioning in the US and is a well-known brand for both consumers and lenders.
  
  
• Mission-critical: Its product is a critical input in the lending/securitization process yet accounts for a tiny fraction of the customer's total costs (~$0.05/score). So difficult to replace and very little incentive to do so even with recent special pricing actions.
  
  
• High margins: FICO owns the algorithm that produces the FICO score, not the underlying data. It licenses its algo to customers - primarily credit bureaus - who run the algorithm against their own consumer data and sell their data + the FICO score to lenders who use it in their underwriting process. 
  
  
• High growth: Beginning in FY18 FICO began implementing 'special pricing' actions to narrow the gap between the value it is providing in the underwriting process vs. the price it was receiving. Since FY17, Scores revenues have increased at a 28% CAGR.
  
  
• Long runway for Special Pricing: Of the ~15bn scores FICO provides, it has taken special pricing on just 'hundreds of millions.' Management recently characterized the business as being in 'inning #1' in terms of being able to realize special pricing.  
  https://twitter.com/b_cap21/status/1438177341507780615?s=20
  
  
• Less cyclical than historical business: FICO scores had tough results in the GFC, but the combination of its (a) open access program (where consumers can view their FICO scores monthly at majority of large banks), (b) special pricing, and (c) a B2C business (principally its myFICO.com credit monitoring business and partnerships with EXPN ) which didn't exist in the GFC but is now 25% of Scores revenues should make it less cyclical in a downturn.   

Catalyst

A realization that the Scores business growth outlook remains robust