Part l: A History of Visa


Visa is one of the biggest companies in the world. Cards bearing the Visa logo are used more than 340 million times every day. And the Visa brand is one of the most-recognized on the planet. Yet unlike other companies of similar size and ubiquity, few people know what Visa does, how they make money, or why they even exist.

To understand, it helps to look at the company’s history.


"A bank is the place for a poor man to put his money so that a rich man can get it when he wants."

At the turn of the century, most Americans found it difficult to borrow money from banks. For a time, this wasn't a big deal. Not when most Americans lived with a set of low-cost goods, largely unchanged for centuries: furniture, clothing, crude farm tools, basic kitchen and household items, perhaps a wagon. But toward the end of the 19th century, the old way of doing things was challenged by new, technologically advanced goods that flooded the American market. These new goods—cars, telephones, sewing machines, refrigerators—had obvious advantages. They enabled people to do more, in less time, and with less effort than previously possible. 

There was just one problem: most people couldn't afford them.


Merchants and manufacturers came up with a solution: installment credit.

For consumers who needed a car, a new set of tools, or a dishwasher—but couldn't afford to pay for it outright—installment credit allowed them to put a little money down, and pay off the remaining debt in small monthly increments. Goods that would've taken years to save for could now be had for a few dollars a month. Almost overnight, the unaffordable became affordable. 

It wasn't just consumers who benefited. Merchants were quick to realize that people are far more likely to buy something when they can pay for it later. So for any retailer or manufacturer looking to increase sales, selling on installment credit was a simple and effective way to do it. The plans were so popular that by 1930, most durable goods—including 65% of cars; 85% of furniture; 75% of washing machines; 65% of vacuum cleaners; 20% of jewelry; and 75% of radio sets—were sold on credit.

But lending money was not without cost: merchants had to asses each individual customer's creditworthiness; bear the risk of late or nonpayment; and shoulder all of the back office headaches and expenses that came with managing thousands of individual accounts—like billing/collection, postage, stationery, and customer service reps. It was cumbersome for customers, too. Every time they wanted to make a purchase on credit, they had to shlep to the merchant, sit down with an associate, hand over their financial and employment history, and fill out stacks of paperwork. 

The process was a pain—and it was why banks avoided making small loans in the first place. That is, except for one bank.


By the 1950s, Bank of America had grown into the largest bank in the country—and it did so by being the one big bank willing to lend to middle-class consumers. “We were always a leader in installment credit. Anything you could buy on time, we financed: student loans, cars, boats, trailers, home loans, personal loans, you name it", recalls Ken Larkin, a lifelong Bank of America executive. “People would come into the bank 4-5 times a year, whenever they needed extra funds. There was vacation. There were back-to-school clothes. There were the holidays. There was tax time. There were medical emergencies." 

Though the bank flourished because of its willingness to lend money to the middle class, it still suffered the same pains as merchants who sold on installment plans: the reams of paperwork; the tens of thousands of salaried loan officers; the record-keeping; the millions of accounts. Bank of America wanted to make this process more efficient—and thus, more profitable. To do that, they turned to the credit card. 


The idea of a credit card was not new. Diner's Club was the first to popularize the concept in 1951, with a general-purpose card that could be used at multiple merchants. The card differed from other available forms of credit in an important way: instead of receiving credit directly from a merchant, customers received a card—and its associated credit—from a third-party organization not owned by the merchant itself. That third party was Diner's Club. 

Diner's Club signed up different merchants to accept their card, and promised to reimburse them for any charges made on it—even if the cardholder never paid their bill. Thus, the merchant received all of the benefits of selling on credit, without any of the headaches (e.g., bookkeeping, billing, and collecting). And customers benefitted too. For the first time, they could use a single card to make purchases at many merchants. Within 12 months, the card had 42,000 members and was accepted by 330 merchants.


The success of Diner's Club put banks on notice. So in 1958, Bank of America decided to launch their own card—but with a twist. Unlike Diner's Club, Bank of America's card wouldn't require customers to settle their bill in full each month. Rather, their card would offer revolving credit, which differed from installment credit in that there was no fixed number of payments. So borrowers could draw as much as they like, whenever they like, for whatever purchase they like. All without ever setting foot in a bank. 

Bank of America viewed their credit card as a way to provide customers with an instant, no-questions-asked personal loan. Their hope was that by giving customers greater control, the bank could cut down on the time and money spent on processing loans.

“It handed the keys to the customers. It was he or she alone who got to make the decisions about how, and when, to spend large sums of money—and how, and when, to pay the money back. It could be used impulsively or carefully; frequently or sparingly; for emergencies or for shopping sprees. And then, when the bill arrived, it was he or she alone who decided whether to pay back the money all at once (with no interest), or in installments (with interest). The crucial point is that the customer was in control of financial decisions that had always before required the explicit approval of a banker or merchant. One major reason previous credit card programs had flopped was that most banks feared giving their customers that much control. Bank of America had no such qualms.” [1]

The next step was getting people to use the card. And merchants to accept it.


"A successful credit card program requires the participation of not just customers but store owners as well. In fact, it requires thousands of store owners, all of whom have to be recruited with the promise that there will be enough cardholders to make accepting the card—and handing over a percent of the purchase price to a bank—worth their while. At the same time the bank is recruiting merchants, though, it must also recruit cardholders—promising them that there will be enough merchants signed up to make carrying the card worth their while. It's a chicken-or-egg dilemma. What comes first, the customers or the merchants?" [1]

Bank of America's solution to this problem was The Fresno Drop. The bank decided that rather than try to sign up new customers, they would pre-approve existing Bank of America customers living in Fresno, CA. Why Fresno? Because 45% of the city's families were Bank of America customers; and because if it flopped, Fresno's relative isolation would limit the damage to the bank's reputation. With so many of the city's residents set to become cardholders practically overnight, merchants had no choice but to take notice. And so they began to sign on. 

"It was the smaller merchants who first came around. Larkin remembers visiting a drugstore, hoping to persuade its owner to accept BankAmericard. "When I explained the concept of our credit card," he says, "the man almost knelt down and kissed my feet. 'You'll be the savior of my business,' he said. We went into his back office," Larkin continues. "He had three girls working on bookkeeping machines, each handling ~ 1,500 accounts. I looked at the size of the accounts: $4.58; $12.82. And he was sending out monthly bills on these accounts. Then the customers paid him maybe three or four months later. Think of what this man was spending on postage, labor, envelopes, stationery! His accounts receivables were dragging him under." [1]

On September 18, 1958, sixty thousand unwitting Fresno residents opened their mailboxes to find the first true credit card: the “BankAmericard”. Within a year the card was expanded to Los Angeles, San Francisco, Sacramento, Bakersfield, and Modesto. It was a disaster. Fraud was rampant. 22% of cardholders were late on payment. Large merchants were resistant to the idea of paying a percentage of each sale to Bank of America. After 15 months, the program lost $20m (about $170m today). 

Other banks, seeing the early losses of Bank of America's program, stayed away. But this was fortuitous—because by the time Bank of America turned their program profitable three years later, it was too late for competitors to catch up. The BankAmericard had already blanketed the entire state of California. And then the virtuous cycle began: year after year, more cardholders were spending more money on their BankAmericards. To capture those sales, more merchants began to participate in the program. And the more merchants that accepted the card, the more customers used it. It was a cycle that rival banks would be hard pressed to replicate. After all, what customer is going to use a card that isn't accepted by most merchants; and what merchant is going to accept a card that no one uses?


By the mid-1960s, the BankAmericard program had overcome its initial difficulties and was generating solid profits. But because Depression-era bank laws forbade banks from having out-of-state customers, the BankAmericard could only be used in California. So in 1966, to grow the card program beyond their home state, Bank of America decided to franchise the BankAmericard to select banks all over the world. Under the program, Bank of America charged each licensee bank a $25,000 entry fee, plus a royalty of ~0.5% on total cardholder spend. In return, Bank of America would show the smaller bank how to run a credit card operation. 

Merchants that were signed up by the licensee banks agreed to accept all BankAmericards, allowing consumers to use their BankAmericards at any participating merchant worldwide—regardless of which bank the merchant used. In short, the program enabled merchants and customers to have different banks for the first time. 

This was a big deal because now a customer from Albany could get a BankAmericard from their local bank; and they could use that card to buy something from a merchant affiliated with a bank in Hawaii. It looked like a win-win: merchants exponentially expanded their pool of available customers, and customers got to use their card at a growing number of merchants.


But almost as soon as it launched, the licensing program began to fall apart.  

The first problem was authorization. Authorization is what makes it possible for a merchant to know that a customer's card is valid (i.e., not stolen), and that he isn't exceeding his credit limit. Authorizing a transaction is relatively easy when the merchant and customer belong to the same bank: the merchant simply calls their bank to see if the customer's account is in good standing, and the bank looks up the cardholder's account. But with the franchise system, the customer and merchant often had different banks. This made authorizing a transaction much more complicated. 

"The merchant had to call his bank, who then put the merchant on hold while they called the cardholder's bank. The cardholder's bank then put the merchant's bank on hold while they pulled out a big printout to look up the customer's balance to see if the purchase could be approved—all while the customer and merchant stood there in the store, waiting for the reply. And that was when the system was operating smoothly. Sometimes the merchant got a busy signal. Other times his calls went unanswered, something that happened most often when a customer with a card from an East Coast bank tried to buy something late in the day in California. And if the merchant couldn't get through, they then had to decide whether to accept the card or lose the sale. And if the merchant took the card, and it turned out to be stolen, all hell would break loose as the banks fought over who should absorb the losses." [1]

The second problem was the behind-the-scenes system—designed to facilitate transactions between banks—known as interchange

"It is impossible to overstate the importance of a workable interchange system; without it, nationwide credit cards simply cannot exist. Because the vast majority of credit card transactions involve two separate banks—one that handles the merchant's business, and another that issues the credit card to the customer—banks have to have a way that allows them to “settle” their accounts with each other." [1]

When a Bank of New York cardholder bought something at a merchant who was affiliated with Bank of Hawaii, Bank of Hawaii had to have a way to get reimbursed by Bank of New York. This may not sound like such a big deal, but remember, this was before computers were widely used. So in order to settle with one another, the merchant bank had to physically mail cratefuls of paper and sales receipts to customer banks all over the country, on an almost-daily basis. The cardholder banks then had to manually match up the sales drafts with their customer's accounts, reimburse the merchant's bank, and then finally bill the cardholders. 

"It was one thing to settle accounts with five banks or even twenty-five. It was another thing to settle accounts with 150 banks, with millions of cardholders, billions of dollars in sales, and to do it without computer help, with sales drafts flying back and forth across the country every day, and with balances that didn't add up correctly half the time. Back rooms filled with unprocessed transactions, customers went unbilled, and suspense ledgers swelled like a hammered thumb." [1][2]

Bank of America had created their credit card to cut down on the operational headaches of lending money. But in an ironic twist, their decision to license the BankAmericard became an accounting nightmare of its own. As more banks signed on, the system began to break down.


As licensee banks drowned in paperwork and red ink, they threatened to abandon the fledgling BankAmericard system. So in October of 1968, Bank of America summoned them to Columbus, Ohio for a meeting to sort out the technical problems with authorization and interchange. After a tense few days, Bank of America was persuaded to spin off the BankAmericard program into its own standalone entity, co-owned by all member banks. Banks would still compete for merchants and cardholders, but they agreed to cooperate at the card system level by setting operational standards and policy, developing infrastructure, sharing advertising costs, and building technology.

The newly-independent National BankAmericard Inc (“NBI”) was given a mandate by its member banks: build the behind-the-scenes apparatus that would enable banks to run their credit card operations smoothly and profitably. They quickly got to work. In 1973, NBI used computers to automate the authorization process. Three years later, they did the same to interchange. The “digitization” of these processes immediately bore fruit: transactions could now be processed 24/7/365; authorization times dropped from 5 minutes to 50 seconds; banks cleared and settled transactions overnight (instead of a week or more); and postage and labor costs were reduced by $17m ($79m today) in the first year alone. 

NBI saved the program by taking a manual process—calling banks, looking up accounts, mailing sales drafts and receipts, transferring funds—and automating it. With the system finally on firm footing, National BankAmericard Inc. decided to rebrand. The name they chose was Visa.

To be continued…

Sources/Further Reading:

  1. A Piece of the Action: How the Middle Class Joined the Money Class, by Joe Nocera

  2. One from Many: VISA and the Rise of Chaordic Organization by Dee Hock

  3. VISA: The Power of an Idea by Paul Chutkow

  4. Financing the American Dream by Lendol Calder

  5. Payments Industry Overview by @find_me_value

  6. Clearing and Settlement of Interbank Card Transactions: A MasterCard Tutorial for Federal Reserve Payments Analysts by Susan Herbst-Murphy

  7. Mastercard - Value Investors Club by olivia08

  8. Paying with Plastic: The Digital Revolution in Buying and Borrowing by David S. Evans

  9. Payments Systems in the U.S. - Third Edition: A Guide for the Payments Professional by Carol Coye Benson

  10. Electronic Value Exchange:Origins of the VISA Electronic Payment System by David L. Stearns

The Resilience of Costco

For 40 years, Costco has succeeded with a simple formula: reinvest merchandising profits into lower prices and better products; be a disciplined operator; and treat customers and employees well. 

But with greater share of shopping moving online, it’s fair to wonder if the company's best days are behind it. The deck below will explore this in detail. 

A Conversation with Charlie Munger and Michigan Ross Dean Scott DeRue

[Click here to download a PDF copy]

SCOTT DERUE: As you look back on your life experience, what's the most important piece of advice that you would offer everyone in the room tonight as they look forward and into their futures?

MUNGER: Well, there are a few obvious ones; they're all ancient. Ben Franklin, marriage is just like the most important decision you have and not your business career. It'll do more for you—good or bad—than anything else and Ben Franklin had the best advice ever given on marriage. He said, "Keep your eyes wide open before marriage and half shut thereafter”.

It's amazing how if you just get up every morning and keep plugging and have some discipline and keep learning, and it's amazing how it works out okay. And I don't think it's wise to have an ambition to be President of the United States or a billionaire or something like that because the odds are too much against you; much better to aim low. I did not intend to get rich. I just wanted to get independent; I just overshot.

Read More

Viad (VVI)


Since work on the 13F Screener has kept me busy, I haven't had time to do a formal writeup of my most-recent investment—Viad Corp (VVI). While only a 1.5% position, it's a good company trading at a very reasonable price, and I may buy more. But rather than do a full writeup, this time I'm posting a:

  1. VVI "Living Document" (running overview of operations, financials, management, valuation, etc.)
  2. VVI Valuation Model (pro forma financials, DCF, margin analysis, etc.)

Whenever I research a new company, I'll start by creating these two documents, which I then edit and add to as events and circumstances change. For the Living Document, I'll add links to other analyst write-ups, interesting reports or articles about the company/industry, interviews with management, and investor presentations. I'll also include a running list of highlights/lowlights from conference calls,10-Ks, and 10-Qs to see if management actually does what they say they're going to do. 

For the Valuation Model , I create historical and pro forma financial statements, making conservative adjustments and estimates. Then I'll run a few models (e.g., DCF, Residual Earnings, EV/EBITDA) at various margin and growth rate assumptions. The goal here isn't to come up with an exact share price or multiple, but rather to see what range of values are likely, and if the current price affords a healthy margin of safety. 

For Viad, I've published both documents below; alternatively, the Google Docs versions can be accessed via the links above. 

Read More

What's Amazon's Plan for Whole Foods?


It’s clear that Amazon’s $13.7b acquisition of Whole Foods will result in lower prices at the notoriously expensive grocer. What’s less clear is how Amazon will accomplish this. Most expect them to bring-to-bear their formidable supply chain and delivery know-how, driving prices lower by revolutionizing how groceries are shipped and distributed. While these are surely some of Amazon’s long-term goals, this presupposes that Amazon has already solved the unique challenges of storing and shipping groceries. The reality is Amazon does not yet have the core competencies required to operate a grocery supply chain. These they will learn from Whole Foods. 

But if improvements to the supply chain or logistics are still a ways off, then the question remains: how will Amazon reduce prices? We posit that by automating in-store operations via technology (e.g., self-checkout) and pushing more private label products, Amazon will drive down expenses by billions of dollars. They will then pass these savings right back to the customer in the form of lower prices and better value for Prime members. 


Grocery is the holy grail of retail—nearly every individual must buy groceries on a regular basis, and what does get purchased is often the same from week to week (e.g., meat, eggs, milk, bread). At $650b, annual US grocery sales are 6x Amazon’s revenue, yet only 2% of those sales are done online. So grocery presents a massive opportunity for Amazon, but despite the company’s operational efficiency and extensive distribution network, they have not been successful in this space. Why? 

Look no further than the unique challenges of selling groceries. Amazon may have mastered shipping uniform, non-perishable, and scannable items like books and electronics, but shipping goods that spoil, have inconsistent availability and fluctuating costs, require refrigeration, and varying quality of individual units, is a significantly more difficult challenge. Had Amazon already solved this, they wouldn't have needed Whole Foods in the first place. 


As Amazon experiments with logistics and distribution, their near-term strategy is likely to focus on three areas: improve in-store operations via technology and automation; introduce more private label goods like AmazonBasics and 365; and increase the value of Prime. 

1) In-store Operations 

In January, Amazon revealed AmazonGo, which is a suite of technologies (e.g., cameras and sensors) that enable customers to walk into a store, grab what they want, and walk out—all without needing to wait in line or see a cashier. By automatically detecting what a customer took off the shelf, Amazon can bill their account, monitor in-store inventory levels, and reduce “shrink” (i.e., theft). We expect AmazonGo technology to be pushed out across Whole Foods’s 465 stores. 

If we assume half of Whole Foods’s 100k employees are cashiers—and that their total hourly cost to the company (including wages, benefits, perks, etc.) is $25/hour—then Whole Foods likely spends around $2.5b/yr on this expense—13% of total sales. 

2) Private Label Goods 

No longer seen as cheap, low-quality knockoffs, private label goods are now sought after by savvy customers because of their attractive value proposition: same quality, lower price. Brands like Kirkland, Trader Joe’s, 365, and AmazonBasics, are going toe-to-toe with expensive name brand alternatives; and in many cases they’re winning. And it’s not just customers who benefit. By selling private label goods, companies earn higher margins, gain product exclusivity, and have greater control over packaging and distribution. Since Amazon and Whole Foods have been independently selling more low-priced, high-margin private label goods, we expect the acquisition to accelerate this trend. 

By improving in-store operations and selling more private label goods, Amazon stands to increase the profitability at Whole Foods. But rather than keep the profit, Amazon has other plans…

3) More Value for Prime Users

We expect Amazon to “give back” any profits to customers through lower prices and increased value for Prime members. To understand why, look no further than the company that serves as Prime’s inspiration: Costco.

Costco sells products “at cost”, meaning if it costs them $1 to supply customers 1lb of turkey, then that’s the same price they charge those customers. While this virtually guarantees Costco can offer the lowest prices, it also seemingly guarantees they’ll never make money. But there’s a catch—to get Costco’s low prices, customers must pay $60/year to be a “member”. It’s a unique but brilliant business model; customers win because they receive more than $60 worth of savings each year, and Costco also wins because there’s no incremental cost to provide a membership, so all membership fees fall directly to their bottom line.

This is the same model that Amazon Prime means to replicate. Since the company doesn’t need to make money directly off the goods it sells, it can keep prices lower than its competitors. The low prices drive more memberships, which increases sales, which in turn gives the company more leverage to get even lower prices from its suppliers. Rather than keep that margin, the company gives it right back to the members through lower prices, thus driving more memberships and restarting the virtuous cycle.

Because of the acquisition, Amazon will have a number of new opportunities to increase Prime’s value, such as: free grocery delivery, member-only sales/pricing, exclusive brands, online ordering/pickup, and easy re-ordering. And by increasing Prime’s value, Amazon attracts more Prime members, which increases sales, which in turn gives the company more leverage with suppliers, which drives prices lower, which….well you get the point. 


While predictions of drone-delivered swiss chard may make for good headlines, the future success of Whole Foods does not depend upon it—nor any other single technology for that matter. What it will depend on is whether or not Amazon can create value for customers through low prices, great selection, and convenience. Over the next few years, we think it is far more likely for Amazon to achieve these goals through slightly less exciting—though no less impactful—methods like self checkout technology and selling more private label goods. Both changes will substantially reduce costs in the near-term, which Amazon will use to accelerate the virtuous cycle of AmazonPrime.