S-1 Teardown of Affirm

• First interesting thing to notice is that there are only a few bankers on this deal (9 in total)

• Compare this to Airbnb (38 banks!) and Doordash (12) which are both consumer facing products recently going public - more firms typically means bigger capital raises

• Allen and Company is a joint bookrunner on AFRM

  • Allen and Co is a famous investment bank that typically only deals with high value public companies (Dara Khosroshawi the Uber CEO was a Managing Director at Allen and Co.)

  • It buys a significant amount of shares in companies when they go public and typically holds the shares themselves rather than placing it with high net worth clients like Goldman or JP Morgan

• "With most of the payments industry deriving profits from late fees, overdraft charges, and gimmicks like deferred interest, it's not hard to agree that there has to be a better way: it's time to evolve payments again. We founded Affirm in 2012 to lead this change, and to build the first payment platform with a moral backbone and consumer-first mindset."

  • I may be crazy but high interest fees are a revenue source for Affirm right?

• "Our core strengths and competitive advantages remain in two fundamental areas: we are very good at technology, and we are committed to being good."

  • Is technology a competitive advantage?

  • Why is "being" italicized?

• Mission - "Deliver honest financial products that improve lives."

  • DO FINANCIAL PRODUCTS IMPROVE LIVES? Maybe?

Products and Market

• "Today, it comprises three core elements: a point-of-sale payment solution for consumers, merchant commerce solutions, and a consumer-focused app."

• According to eMarketer, global online sales grew 20% to approximately $3.4 trillion in 2019 and are expected to grow to approximately $5.8 trillion by 2023; however, e-commerce still only accounts for 14% of total retail sales.

  • LOL eMarketer reports are always spam hidden behind paywalls; I wonder what it actually says

• "In the United States, according to the U.S. Department of Commerce, e-commerce sales as a percentage of total sales jumped from 11.8% to over 16.1% between the first and second calendar quarters of 2020."

  • These stats I do trust and that is a substantial and wild jump

  • The question is, will it stay like that?

• According to Worldpay’s 2020 Global Payments report, “buy now pay later” is the fastest growing e-commerce payment method globally. In North America, “buy now pay later” market share is expected to triple to 3% of the e-commerce payments market by 2023. In other regions, such as EMEA, “buy now pay later” already accounts for almost 6% of the e-commerce payment market, and is expected to grow to almost 10% by 2023.

  • Interesting commentary on EMEA where Klarna is definitely more competitive and prevalent - Affirm betting on similar traction in the US over time

• "In 2019, consumers paid approximately $121 billion in credit card interest, $11 billion in overdraft fees, and $3 billion in late fees according to studies from LendingTree, the Center for Responsible Lending, and NerdWallet, respectively."

  • Affirm's messaging is all about how bad late fees and overdraft fees are but its really credit card interest (something which Affirm does make money on - well not credit specifically but interest, yes) that is really a big big problem

• "Debit cards restrict buying power to a consumer’s account balance, constrain the ability to budget over time, and do not provide the optionality or ability to responsibly access credit."

  • Some people would say that is a positive for consumers - not spending more than they currently have

Value Proposition

• More customers and better conversion. Our merchants report increased customer conversion when they offer Affirm. Our dollar-based merchant retention has consistently exceeded 100% for each cohort of merchants that joined our platform since 2016. According to a survey conducted by Informa Business Intelligence from November 1, 2019 to January 31, 2020, Affirm also approves, on average, 20% more customers than comparable competitor products.

  • So basically the companies who add Affirm do more business with Affirm over time

    • Ex: 100% means if they had $10 of sales in 2019 with Affirm they had $10 of sales again in 2020. 110% would mean $11.

• Increased AOV. By solving affordability for consumers, we are able to help merchants increase demand for higher net average order value (“AOV”) items. We calculate AOV as our GMV divided by the number of transactions conducted on our platform. According to internal studies we conducted in 2018 and 2019 comparing AOVs before refunds, merchants using Affirm reported 92% and 85% higher AOVs, before refunds respectively, when compared to other payment methods.

  • The before refunds is an interesting caveat - Affirm has to increase impulse purchasing and thus I'd imagine higher refunds?

  • I'm confused by this stat - Did 92% and 85% of merchants record an increase in AOV or did the AOV itself increase 92%? Those would be drastically different outcomes

  • Also citing internal surveys is kinda weeeeeaaaakkkkkk

• Increased repeat purchase rate. We believe that the frictionless consumer experience and enhanced buying power facilitated by our platform often lead to improved repeat purchase rates for our merchants as they enjoy more satisfied customers. For example, approximately 22% of Affirm consumers from our January 2018 cohort and approximately 20% of Affirm consumers from our January 2019 cohort made repeat purchases at the same merchant within 12 months.

  • This is a big deal for companies

    • Side note: when are we gonna see BNPL for hundred thousand dollar SaaS deals - oh wait its called payment terms and its existed for a long time! 30 - 90 days depending on industry

• Better data to inform personalized promotional strategies. We provide merchants with valuable data they can use to offer more tailored promotions and offers to consumers, for example through targeted offers in our app.

  • Aren't we like not excited about our data being used to funnel us to purchase more things? Like aren't we mad at FB and Google?

  • This makes tons of sense strategically but is Affirm really helping people by increasing their purchase rate and impulse purchasing? Unclear.

• Broader target market. We enable merchants to achieve incremental sales and expand their target market by including customers who might not want to or be able to pay for an item upfront, but can do so responsibly over time.

• Ease of integration. Our direct API, designed for use by developers, allows for site integration with minimal investment. Merchants can easily incorporate our platform into payment and product pages, and we provide a dedicated integration team to ensure that any issues are resolved.

• Compliant. Once a merchant has integrated our API, we handle the regulatory aspect of the loans facilitated through our platform, irrespective of state or jurisdiction.

Competitive Advantages

How Affirm Makes Money

• "From merchants, we earn a fee when we help them convert a sale and power a payment. Merchant fees depend on the individual arrangement between us and each merchant and vary based on the terms of the product offering; we generally earn larger merchant fees on 0% APR financing products. For the fiscal year ended June 30, 2020 and for the three months ended September 30, 2020, 0% APR financing represented 43% and 46%, respectively, of total GMV, facilitated through our platform."

  • I wonder what the merchant fee rates are and how they have trended over time

  • One BIG RISK is that commoditization in the space via Klarna, Afterpay and others means this part of revenue keeps going down as companies offer it cheaper and cheaper

• "From consumers, we earn interest income on the simple interest loans that we purchase from our originating bank partners. Interest rates charged to our consumers vary depending on the transaction risk creditworthiness of the consumer, the repayment term selected by the consumer, the amount of the loan, and the individual arrangement with a merchant. Because consumers are never charged deferred or compounding interest, late fees, or penalties on the loans, we are not incentivized to profit from our consumers’ mistakes or misfortunes."

  • So 44% of revenue comes from simple interest loans that they purchase from bank partners - who originate said loans

    • Their bank partner is Cross River Bank - a small bank in New Jersey that has focused on leveraging the changing fintech ecosystem - something that smaller banks are able to do better because they face less restrictions

• "As of September 30, 2020, we had over $4.2 billion in funding capacity from a diverse set of capital partners, and we have funded approximately $10.7 billion of purchases since July 1, 2016."

  • At any given time, through its financing partners, Affirm users can buy $4.2B worth of stuff - WOW

• The company has worked with 6.2M consumers and has 6,500 merchants integrated on its platform

  • The number of merchants seems a bit low maybe? They recently partnered with Shopify as an additional route for merchant growth

    • One question is - why is there no self-service merchant on-boarding - like stripe accepting payments super easily - affirm could help merchants do BNPL super easily

      • There is probably a regulatory hurdle to this

Growth Strategies

• Expand To More Higher Frequency Purchases

  • This one is really interesting because higher frequency purchases typically cost less but the market is obviously huge

  • Think about things that could be bought this way that aren't right now - groceries, subscriptions, toiletries - this is kind of what Afterpay has done in terminal purchases in Australia

    • If a consumer wants an ice cream, but doesn't have the cash at the convenience store - let them purchase the ice cream using Afterpay/Affirm and pay it back with a tiny amount of interest over time

    • Does the store then have a working capital issue (getting rid of the inventory and not receiving the cash) or does Affirm front the payment to the store and then make the small small amount of interest on the back end? Not sure but either way its something that has little traction so far in the US

  • THIS IS SORTA PAYDAY LENDING AT BETTER TERMS THOUGH RIGHT?

• Expand Consumer Reach

  • BIG SURPRISE

• Expand Merchant Reach

  • ANOTHER BIG SURPRISE

• Expand to New Markets

  • They don't call out any markets specifically - geographically or software solutions

Risk Factors

• "Our top merchant partner, Peloton, represented approximately 28% of our total revenue for the fiscal year ended June 30, 2020 and 30% of our total revenue for the three months ended September 30, 2020. Our top ten merchants in the aggregate represented approximately 35% of our total revenue for the fiscal year ended June 30, 2020 and approximately 37% of our total revenue for the three months ended September 30, 2020."

  • PELOTONNNNNN - this is some pretty significant concentration given the size of the company but thats what happens during a pandemic

  • "In addition, a material modification in the financial operations of any significant merchant partner could affect the results of our operations, financial condition, and future prospects. For example, the timing of our revenue recognition is tied to when a merchant captures payment and confirms a transaction financed through our platform, which we refer to as the merchant capture date. If a merchant recognizes the payment collection and confirms the transaction later in their transaction process, we expect that this change would delay the merchant capture date, which would delay our recognition of GMV and revenue related to that merchant’s transactions by a corresponding amount. Such a delay would adversely affect the GMV and revenue that we recognize from such merchant’s transactions in the quarterly period of such change, as the merchant capture date for a portion of such transactions would shift to a subsequent quarterly period. The implementation of such a change began with respect to Peloton in December 2020. This change will delay the merchant capture dates of certain transactions, which will correspondingly delay the GMV and revenue related to these transactions. This implementation in December 2020 will negatively impact our results of operations, primarily in our second and third fiscal quarters of 2021."

    • So Peloton is now recognizing revenue later in their transactions - I'm gonna bet when the bike gets delivered (not purchased) - given the delays in getting people bikes during this crazy time; this would push out some big transactions from the holiday season

    • I wonder if this was a reason for IPO delay - like maybe Affirm saw a ton of volume from Peloton in early december and said lets wait til the new year when the revenue starts rolling in - more likely it was just a timing thing with the ways investors work in late December (Hint: THEY DONT)

• "We operate in a highly competitive and dynamic industry. Our technology platform faces competition from a variety of players, including those who enable transactions and commerce via digital payments. Our primary competition consists of: legacy payment methods, such as credit and debit cards, including those provided by card issuing banks such as Synchrony, J.P. Morgan Chase, Citibank, Bank of America, Capital One, and American Express; technology solutions provided by payment companies such as Visa and MasterCard; mobile wallets such as PayPal; and other pay-over-time solutions offered by companies such as Afterpay and Klarna, as well as new pay-over-time offerings by legacy financial and payments companies, including those mentioned above."

  • Lot of big, well capitalized companies

• "We rely on Cross River Bank to originate a substantial majority of the loans facilitated through our platform and to comply with various federal, state, and other laws. Cross River Bank handles a variety of consumer and commercial financing programs. The Cross River Bank loan program agreements have initial three-year terms ending in November 2023, which automatically renew twice for successive one-year terms unless either party provides notice of non-renewal to the other party at least 90 days prior to the end of any such term."

  • I guess this is less worrying for Affirm and more worrying for Cross River Bank - which is a small, non-public NJ bank.

• "Changes in market interest rates could have an adverse effect on our business"

  • Increased interest rates may adversely impact the spending levels of consumers and their ability and willingness to borrow money. Higher interest rates often lead to higher payment obligations, which may reduce the ability of consumers to remain current on their obligations and, therefore, lead to increased delinquencies, defaults, consumer bankruptcies and charge-offs, and decreasing recoveries, all of which could have an adverse effect on our business.

    • Interesting point of view here - with the FED issuing guidance that we are going to be in a low interest environment for a while that's better for Affirm

• "In connection with our securitizations, warehouse credit facilities, and forward flow agreements, we make representations and warranties concerning the loans financed pursuant to such agreements. If those representations and warranties are not correct, we could be required to repurchase certain of such loans. Any significant required repurchases would have an adverse effect on our ability to operate and fund our business."

  • Warehouse lending can most simply be understood as a means for a bank or similar institution to provide funds to a borrower without using its capital. A small or medium-sized bank might prefer to use warehouse lending and to make money from origination fees and the sale of the loan rather than earn interest and fees on a 30-year mortgage loan. In warehouse lending, a bank handles the application and approval of a loan but obtains the funds for the loan from a warehouse lender. When the bank then sells the mortgage to another creditor in the secondary market, it receives the funds that it then uses to pay back the warehouse lender. The bank profits through this process by earning points and origination fees.

    • Ok so basically - a warehouse loan is a loan to a loan originator - so people are loaning Affirm to loan other people? I think. Honestly not sure

    • Basic takeaway is this is a DEBT BUSINESS - there is lending, securitization, and lots of dependency on stable rates of return from various parties

• "If our collection efforts on delinquent loans are ineffective or unsuccessful, the performance of the loans would be adversely affected."

  • This is still a loan sharky business lol

  • In addition, because our servicing fees in connection with the services we provide depend on the collectability of the loans facilitated through our platform, if there is an unexpected significant increase in the number of consumers who fail to repay their loans or an increase in the principal amount of the loans that are not repaid, we will be unable to collect our entire servicing fee for the loans facilitated through our platform for which we act as servicer, and our business, results of operations, financial condition, future prospects, and cash flows could be materially and adversely affected.

    • Isn't this ummmm against your beliefs of bringing honesty to financial products?

• "We identified a material weakness in our internal control over financial reporting in connection with the audit of our financial statements for the fiscal year ended June 30, 2020, and we may identify additional material weaknesses in the future that may cause us to fail to meet our reporting obligations or result in material misstatements of our financial statements. If we fail to remediate any material weaknesses or if we otherwise fail to establish and maintain effective internal control over financial reporting, our ability to accurately and timely report our financial results could be adversely affected."

  • The material weakness relates to our general information technology controls, including design and implementation of access and change management controls. This material weakness means that it is possible that our business process controls that depend on the affected information technology systems, or that depend on data or financial reports generated from the affected information technology systems, could be adversely affected. Key components of the COSO framework have not been fully implemented, including control and monitoring activities, and information and communication relating to: (i) electing and developing general control activities over technology to support the achievement of objectives; (ii) electing, developing, and performing ongoing and/or separate evaluations to ascertain whether the components of internal control are present and functioning; and (iii) generating and using relevant, quality information to support the functioning of internal control.

  • Ok so basically they didn't have the proper governance and controls in place to prevent all employees from accessing confidential financial information

  • Having a material weakness in IPO is not great - it means that the Company didn't have its shit together to go public and kinda makes you consider whether the CFO knew

  • At least its not revenue related but rather governance controls

• "Determining our allowance for credit losses requires many assumptions and complex analyses. If our estimates prove incorrect, we may incur net charge-offs in excess of our reserves, or we may be required to increase our provision for credit losses, either of which would adversely affect our results of operations."

  • This is what ultimately did the banks in during the financial crisis - their provision for losses was too low and they couldn't handle the major real estate losses

How Does Affirm actually work?

GMV Based Cohort Performance

COVID-19 Impact

Revenue Categories Shmategories

• Merchant network revenue. Merchant partners are charged a fee on each transaction processed through the Affirm platform. The fees vary depending on the individual arrangement between us and each merchant and vary based on the terms of the product offering. The fee is recognized as earned when the terms of the executed merchant agreement have been fulfilled and the merchant successfully confirms the transaction. In each of the fiscal years ended June 30, 2019 and 2020, we generated 50% of our revenue from merchant network fees. During the three months ended September 30, 2019 and 2020, we generated 41% and 54% of our revenue, respectively, from merchant network fees.

  • Ok so basically no transparency on any fee arrangements nor the revenue recognition policies - hmmmm

• Virtual card network revenue. A smaller portion of our revenue comes from our Virtual Card product. We partner with issuer processors for the creation of virtual debit cards used by customers at checkout. Consumers can apply at affirm.com or through the Affirm app and, upon approval, receive a single-use virtual debit card to be used for their purchase online or offline at a non-integrated merchant. The merchants are charged interchange fees for virtual debit card transactions by the issuer processors, as with all debit card purchases, and the issuer processor shares a portion of this revenue with us. This revenue is recognized as a percentage of both our loan volume transacted on the payment processor network and net interchange income, and this revenue is presented net of associated processing fees. In the fiscal years ended June 30, 2019 and 2020, we generated 3% and 4% of our revenue, respectively, from virtual card network fees. During the three months ended September 30, 2019 and 2020, we generated 4% and 3% of our revenue, respectively, from virtual card network fees.

  • The payments industry loves virtual cards because they are basically the highest margin product in existence; it takes almost no money to make them because they are virtual and the security is also better

  • This is small but could be a cash generating growth area in the future

• Interest income. We also earn revenue through interest earned on the loans purchased from our originating bank partner. Interest income includes interest charged to consumers over the term of the consumers’ loans based on the principal outstanding and is calculated using the effective interest method. In addition, interest income includes the amortization of any discounts or premiums on loan receivables created upon purchase of a loan from our originating bank partner. Discounts and premiums are accreted or amortized over the life of the loan using the effective interest method. In the fiscal years ended June 30, 2019 and 2020, we generated 45% and 37% of our revenue, respectively, from interest income. During the three months ended, September 30, 2019 and 2020, we generated 46% and 31% of our revenue, respectively, from interest income.

  • We don't know average interest rate!

• Gain (loss) on sales of loans. We sell a portion of the loans we purchase from our originating bank partner to third-party investors through our platform. We recognize a gain or loss on sale of such loans as the difference between the proceeds received, adjusted for initial recognition of servicing assets and liabilities obtained at the date of sale, and the carrying value of the loan. During the fiscal year ended June 30, 2019, loss on sale of loans negatively impacted our revenue by less than 1%. During the fiscal year ended June 30, 2020, we generated 6% of our revenue from gain on sales of loans. During the three months ended September 30, 2019 and 2020, we generated 7% and 9% of our revenue, respectively, from gain on sales of loans.

  • They are generating more revenue from sales of loans - this could either be because the quality of the loan portfolio (from a credit score / ability to repay) is improving or because there is more interest from buyers of securitized loans

• Servicing income. We earn a specified fee from providing professional services to manage loan portfolios on behalf of our third-party loan owners. Under the servicing agreements with our capital markets partners, we are entitled to collect servicing fees on the loans that we service, which are paid monthly based upon an annual fixed percentage of the outstanding loan portfolio balance. In the fiscal years ended. June 30, 2019 and 2020, we generated 2% and 3% of our revenue, respectively, from servicing fees. During each of the three months ended September 30, 2019 and 2020, we generated 2% of our revenue from servicing fees. We expect our revenue may vary from period to period based on, among other things, the timing and size of onboarding of new merchants, the mix of 0% APR loans versus interest-bearing loans with simple interest, type and mix of products that our merchants offer to their customers, the rate of repeat transactions, transaction volume, and seasonality of or fluctuations in usage of our platform.

  • This is a minimal part of revenue but interesting because Affirm is not using a third party servicing firm - they are servicing loans on their own

Cost Categories Shmategories

• Loss on Loan Purchase Commitment. We purchase loans from our originating bank partner that are processed through our platform and our originating bank partner puts back to us. Under the terms of the agreement with our originating bank partner, we are generally required to pay the principal amount plus accrued interest for such loans and the loan origination fee. In certain instances, our originating bank partner originates loans with zero or below market interest rates that we are required to purchase. In these instances, we may be required to purchase the loan for a price in excess of the fair market value of such loans, which results in a loss. These losses are recognized as loss on loan purchase commitment in our consolidated statements of operations. These costs are incurred on a per loan basis.

  • This includes the purchase amount plus the small amount of interest generated in the few days before Affirm purchases a loan from their bank partner

  • So its nice how explicitly they say this: We buy 0% loans from our bank partner and we are buying them at a loss

    • This means that if these 0% loans start to not perform at all - then Affirm could have gotten paid a small merchant network fee and not gotten its money back AND bought a loan above market value

• Provision for Credit Losses. Provision for credit losses consists of amounts charged against income during the period to maintain an allowance for credit losses. Our allowance for credit losses represents our estimate of the credit losses inherent in our loans held for investment and is based on a variety of factors, including the composition and quality of the portfolio, loan specific information gathered through our collection efforts, current economic conditions, and our historical net charge-off and loss experience. These costs are incurred on a per loan basis.

  • This is likely highly determined by internal company models

  • This is also subject to significant risk - if loans ever stop performing then this would increase

  • This is a BIG cost (Affirm's second biggest) - there is an inherent tension for Affirm here - the lower we reserve against credit losses the better our profit numbers look

    • They could also do the opposite - continually reserve more money as provision for credit losses to basically break their income statement so they never pay taxes or so they hid profits from Wall Street

• Funding Costs. Funding costs consists of the interest expense we incur on our borrowings, and amortization of fees and other costs incurred in connection with our funding. Excluding the amortization of debt issuance costs, which totaled $1.7 million and $2.3 million for the fiscal years ended June 30, 2019 and June 30, 2020, respectively, and $0.6 million and $1.1 million for the three months ended September 30, 2019 and September 30, 2020, respectively, we incur an expense per loan pledged to our debt funding sources.

  • This is relatively low cost but basically the company is incurring a significant amount of interest expense around its purchasing

  • This should probably be called Interest Expense right?

• Processing and Servicing. Processing and servicing expense consists primarily of payment processing fees, third-party customer support and collection expenses salaries and personnel-related costs of our customer care team, and allocated overhead. Payment processing costs are primarily driven by the number and dollar value of consumer repayments which grow as the number of transactions and GMV processed on our platform increases. Customer care loan servicing costs are primarily staffing costs related to third-party and in-house loan servicing agents, the demand for which generally increases with the number of transactions on our platform. Collection fees are fees paid to agencies as percentages of the dollars of repayment they recuperate from borrowers who had previously charged off. Processing and servicing expenses are predominantly per transaction processing fees and third-party staffing fees that generally increase with consumer contact.

  • This is another smaller cost but does show the cost of support - this could be considered gross margin potentially?

    • Its probably more of Processing and Serving, Loss on Loan Purchase commitment and Funding costs

    • This cost structure layout is weird because there is probably some hosting / serving costs in the Technology expense

• Technology and Data Analytics. Technology and data analytics expense consists primarily of the salaries, stock-based compensation, and personnel-related costs of our engineering and product employees as well as our credit and analytics employees who develop our proprietary risk model, which totaled $53.2 million and $75.9 million for the fiscal years ended June 30, 2019 and 2020, respectively. During the three months ended September 30, 2019 and 2020, these expenses were $17.6 million and $21.1 million, respectively. Additionally, for the fiscal years ended June 30, 2019 and 2020, $12.6 million and $17.1 million, respectively, of salaries and personnel costs that relate to the creation of internally-developed software were capitalized into property, equipment and software, net on the consolidated balance sheets, and amortized into technology and data analytics expense over the useful life of the developed software. This amortization expense totaled $2.9 million and $5.5 million for the fiscal years ended June 30, 2019 and June 30, 2020, respectively. During the three months ended September 30, 2019 and 2020, $4.0 million and $5.1 million, respectively, of salaries and personnel costs that relate to the creation of internally-developed software were capitalized into property, equipment and software, net on our consolidated balance sheets, and we recorded amortization expense of $0.6 million and $2.6 million, respectively. Additional technology and data analytics expenses include platform infrastructure and hosting costs, third-party data acquisition expenses, and expenses related to the maintenance of existing technology assets and our technology platform as a whole.

  • This is a massive cost for Affirm and weird because it includes specific serving related costs (infra and hosting) as well as proprietary software development models

  • The company is also capitalizing its software costs - basically this means recognizing development as an investment rather than an operating expense - this would increase profits because the investment in software development would only occur as the amortized value of the full investment in the income statement

    • Ex: $25M in development may last five years; rather than saying we spent 25M in tech and data analytics, you say we will recognize $5M of expenses ever year

• Sales and Marketing. Sales and marketing expense consists primarily of salaries and personnel-related costs, as well as costs of general marketing and promotional activities, promotional event programs, sponsorships, and allocated overhead. In July 2020, we recognized an asset in connection with a commercial agreement with Shopify Inc. in which we granted warrants in exchange for the benefit of acquiring new merchant partners. This asset represents the probable future economic benefit to be realized over the four-year expected benefit period and is valued based on the fair value of the warrants at the grant date. This value is amortized on a straight-line basis over the four-year period into sales and marketing expense, due to the nature of the expected benefit. We expect that our sales and marketing expense will increase as a percentage of revenue as we expand our sales and marketing efforts to drive our growth, expansion, and diversification.

  • Its weird how small this expense is given what affirm is

  • But at the same time the business has great inherent marketing by having its brand embedded in checkout for companies

• General and Administrative. General and administrative expenses consist primarily of expenses related to our finance, legal, risk operations, human resources, and administrative personnel. General and administrative expenses also include costs related to fees paid for professional services, including legal, tax and accounting services, and allocated overhead. Following the completion of this offering, we expect to incur additional expenses as a result of operating as a public company, including costs to comply with the rules and regulations applicable to companies listed on a national securities exchange, costs related to compliance and reporting obligations pursuant to the rules and regulations of the SEC, and increased expenses for insurance, investor relations, and professional services. We expect that our general and administrative expense will increase in absolute dollars as our business grows.

  • This is somewhat big but reasonable given the complex nature of Affirm's business

• Other Income (Expense), Net. Other income (expense), net consists of interest earned on our money market funds included in cash and cash equivalents and restricted cash, and gains and losses incurred on both our constant maturity swaps and as related to bifurcated derivatives associated with our convertible debt.

Quarterly P&L

Credit Metrics

Recent Events and Securitizations

• On July 1, 2020, Affirm Asset Securitization Trust 2020-Z1, entered into a note purchase agreement with Barclays Capital Inc. to issue $150.0 million of fixed rate asset-backed notes with a maturity date of October 15, 2024. The notes bear interest at a rate of 3.46% per year. On July 8, 2020, we contributed $169.8 million of loan receivables to the securitization. We are required to retain 5% of the fair value of the transaction.

  • This is basically asset based lending - like mortgage back securities - not bad if its not too significant / the loans don't go bad

• On July 16, 2020, we entered into a customer installment program agreement with Shopify Inc. to offer its payment product on the SHOP App. As part of this agreement, we issued warrants for common stock up to an aggregate of 20,297,595 shares. These warrants vest in accordance with the terms of the agreement, which include an acceleration of vesting for the consummation of an initial public offering.

  • The agreement has an initial three-year term ending in July 2023, which automatically renews for additional and successive one-year terms unless either party provides the other party with written notice of election to terminate the agreement at least 180 days prior to the end of any such term. Either party may immediately terminate the agreement after the first year of the initial term without cause by providing at least 180 days’ prior written notice to the other party. In addition, upon the occurrence of certain early termination events, either party may terminate the agreement immediately upon notice to the other party.

  • Until the termination of the agreement, Shopify agrees that we will be its exclusive provider in the United States of a closed-end installment loan product (or any substantially similar financial product or program). Pursuant to the agreement, we will be Shopify’s exclusive provider in the United States and its territories of interest-bearing loan installment programs, contingent upon such programs being mutually developed and approved, and upon negotiation of a mutually acceptable agreement by the parties for such programs.

• On August 5, 2020, Affirm Asset Securitization Trust 2020-A, entered into a note purchase agreement with Barclays Capital Inc., Morgan Stanley & Co. LLC, and Credit Suisse Securities USA LLC to issue $368.3 million of fixed rate asset-backed notes with a maturity date of February 18, 2025. The notes bear interest at rates ranging from 2.10% to 6.23% per year. On the same day, we contributed $404.6 million of loan receivables to the securitization. We are required to retain 5% of the fair value of the transaction.

  • Its interesting they just decided to start doing this in 2020 - especially when the company though it was going to go public - this is raising debt capital

    • Less dilutive than equity - but then they went out and raised equity in September? Just seems odd but I guess it lets Affirm move some loans off of its books and get more capital?

  • The notes under the 2020-A securitization were issued in three classes: Class A in the amount of $330.0 million, Class B in the amount of $16.2 million and Class C in the amount of $22.1 million (collectively, the “2020-A Notes”). The Class A, Class B and Class C notes bear interest at a fixed rate of 2.10%, 3.54% and 6.23%, respectively, and each class has a maturity date of February 18, 2025. Principal and interest payments began in September 2020 and are payable monthly. These notes are recorded at amortized cost on the balance sheet. The associated debt issuance costs of $3.4 million are deferred and amortized into interest expense over the contractual life of the notes. The notes held by third-party investors and the unamortized debt issuance costs are included in notes issued by securitization trusts with a balance of $364.8 million on the consolidated balance sheets at September 30, 2020 and are secured by unsecured consumer loans at amortized cost of $399.2 million included in loans held for investment on the consolidated balance sheets as of September 30, 2020.

    • Ok so we are now traunching these securitizations - getting a little more interesting from a loan issuing perspective

• On September 11, 2020, we issued 21,824,141 shares of Class G redeemable convertible preferred stock for an aggregate purchase amount of $434.9 million. These shares have a liquidity preference of $434.9 million. As part of this equity financing round, the convertible notes issued in April 2020 converted into 4,444,321 shares of Class G-1 redeemable convertible preferred stock.

Cash Flow and Balance Sheet

• This cash flow statement is ummmm DENSE.

  • On the Operating activities, apart from the losses the big items that stand out are the purchases of loans held for sale, the proceeds from sales of loans - this is Affirm buying and selling loans

• The Cash from investing opps is interesting

  • They are buying loans which is costing a lot; then getting interest payments on the loans, selling some loans into securitizations

• Cash from Financing opportunities

  • The company is financing most of its loan purchasing with funding debt

• On the balance sheet - loans held for investment is the biggest and growing part of the company's assets

• Funding debt is a huge liability - its clear they are financing almost all purchases with funding debt - this is not a small amount of funding debt - Affirm is highly levered betting on a tight spread between the merchant fees and interest received

Team, Board, and Compensation

Key Considerations

• I think the biggest consideration for Affirm is its product set - how competitive / differentiated is it really? Compared to Klarna / Afterpay?

  • Afterpay has close to 10M active customers, 55k merchants, 11B in GMV, 502M in revenue from a financial perspective

  • Its certainly got competitive metrics - Affirm has a better take rate but lower GMV, lower customers, and lower merchants.

  • This signals the biggest product risk - IS THIS MARKET COMMODITIZED? WILL COMPANIES OPT FOR AFTERPAY BECAUSE THEY TAKE A LOWER TAKE ON THEIR SERVICE

    • If it is - will Affirm be acquired by Max's alma-mater Paypal when they have a competing service?

• The Peloton concentration is definitely a risk in 2023 - what happens when the agreement comes up and Afterpay and Klarna are half the price?

  • Same for Shopify

• Is BNPL here to stay or only getting a bump from COVID-19 related purchasing challenges with many people in and out of work?

• Can Affirm launch more differentiated and complex payments products?

• How penetrated is the market among big merchants for buying things?

  • Seems high based on my online shopping experience

  • Where else can Affirm find growth if it stops post-COVID?

• This is a lending company and the risk of recession tightening payments is really big here - it is Affirm's core business; if we have 5-6% go-forward unemployment post-pandemic then it will not be a pretty economic picture for everyone; on top of that the number of global deaths is impacting economic projections around demographics

IPO Demand / Outcome

• I would expect it to get a ton of retail demand given Afterpay's performance (up 365% in the last year, trading at 35x Forward Sales) and cult following on Twitter LOLOLOL

• Affirm has raised $1.6B - most recently at a known valuation of $2.6B in Apr 2019 (10x YE Revenue)

• I'm going to bet it goes out around a 15-25x FY20 YE End Revenue - $509M * 15-25x = $7.6B - $12.7B valuation

  • Its hard to price anything in this market right now

• It is entirely possible that it gets a 30x+ multiple on some much bigger forward multiple because people are excited about the growth rate

• Either way I expect this to get blown up on Twitter and in the hedge fund world and pushed up pretty quickly

• I am staying away though - its a really complicated funding model with significant risk on the consumer side - its unclear what the economy looks like in a year still so I'll be on the sidelines at these valuations

  • Additionally like this is not really a tech company, its a lending company to super high value consumers - it just doesn't feel like the multiple is really justified; more a tech enabled service than a tech company but the market seems not to care (See $DASH, $ABNB)

Bull Case

• Affirm is the technology and product leader for buy now pay later and its better unit economics cause it to win out long-term against competition

• The market is still largely greenfield (especially in the US) - Affirm takes most of the market share in the US and continues to grow 50%+ for the foreseeable future

• Post-COVID helps Affirm as people are now used to its buying patterns and e-commerce remains really higher than pre-COVID

  • Affirm lays the groundwork now to benefit from a rush to re-open like airlines, restaurants, etc.

• Affirm launches new and interesting payment products in the US - team buying like Pinduoduo for lower value goods in its app, higher value products like cars, homes, which are a whole different funding world and difficulty, further integrations into convenience stores, restaurants, for POS financing

• Shopify contributes meaningfully to Affirm's growth, reducing exposure to Peloton and adding thousands of merchants really quickly

• Affirm develops an auto-merchant onboarding platform that enables merchants to simply embed payments the way they do with Stripe

• Affirm launches payments products for business "consumers" that may be buying low-value subscription / other goods

Bear Case

• This is a commoditized market - Klarna, Afterpay and Affirm start competing on price and Affirm is the most expensive

• Exclusive agreements run out and are renewed at significantly lower rates - Affirm has to start operating cost consciously and cannot continue to add products to its portfolio

  • Probably an acquisition/merger target at that point; but probably not a great value - see Slack as an example

• Its also important to note what happens to price competition - this will hit Affirm in many ways - first a reduction in merchant network fees, then a reduction in interest income (if companies start competing on rates) - these are compounding effects that crush overall revenue growth while INCREASING provisional costs for loan losses - could be bad

• Post-COVID people opt to buy things on credit cards again, ecommerce shrinks as a % of total sales and Affirm's growth drops off significantly - causing the stock to drop as well

• Changes to the economy have a double whammy effect on Affirm - people buy less and do not repay - loss provisions and loss on purchase of loans increases significantly, trashing profits and making it an unattractive acquisition target

  • This would also have rippling effects on the securitizations they are creating - although don't worry this is not a big enough industry to single handedly take down the financial system. Still bad tho