V - Visa

[Liberty]

Acquisition:

 

https://www.reuters.com/article/us-earthport-m-a-visa/visa-to-buy-british-payments-firm-earthport-for-about-250-million-idUSKCN1OQ0EC

 

Visa is paying 198 million pounds ($250.6 million) to buy Earthport Plc (EPO.L), a British firm that facilitates international transactions for banks and businesses, the U.S.-based payments group said on Thursday. [...]

 

Earthport says it offers a lower-cost alternative to traditional payments systems by allowing banks and money transfer firms to have a single relationship instead of multiple ties with various payments channels around the world.

 

For Visa, cross-border payments, or transactions that involve parties in two or more countries, represents a growing business. The volume of such payments rose 10 percent in the 2018 fiscal year, Visa said in October.

[Liberty]

Q4:

 

https://s1.q4cdn.com/050606653/files/doc_financials/2019/q1/Visa-Inc.-Q1-2019-Financial-Results-Presentation.pdf

 

Net income of $3.0B or $1.30 per share

 

Net revenues of $5.5B, an increase of 13%

 

Adj. EPS +21%

 

Underlying business drivers remained healthy: continued double-digit growth in payments volume and processed transactions; cross-border growth slowed

 

Returned $2.9B of capital to shareholders in the form of share repurchases and dividends

 

The board of directors authorized a new $8.5 billion share repurchase program

 

Operating margin: 68%

 

Cash, cash equivalents and investment securities of $15.9 billion

 

Adjusted free cash flow of $3.1 billion for the fiscal first quarter

 

Capital expenditures of $157 million during the fiscal first quarter

[Liberty]

https://www.reuters.com/article/us-earthport-m-a-visa/visa-boosts-offer-for-earthport-in-battle-with-mastercard-idUSKCN1PX0KW

 

Visa sweetened its offer for payment company Earthport Plc to about 247 million pounds ($319.84 million), pushing past rival Mastercard Inc’s earlier bid and setting up a showdown for the assets.

 

Earthport shares have risen more than five-fold to 38.80 pence since Visa first offered to buy the company.

 

Earthport, which had backed the Mastercard bid, said it was now recommending that shareholders accept the increased Visa offer.

[Liberty]

https://www.reuters.com/article/us-paymentprocessors-fees/visa-mastercard-mull-increasing-fees-for-processing-transactions-wsj-idUSKCN1Q41ME

 

Visa Inc and Mastercard Inc, the two biggest U.S. card networks, are preparing to increase certain fees levied on U.S. merchants for processing transactions that will kick in this April, the Wall Street Journal reported on Friday, citing people familiar with the matter.

 

Some of the changes relate to so-called interchange fees, the report said. Interchange fees are what merchants pay to banks when consumers use a credit or a debit card to make a purchase from their store.

 

Fees that Mastercard and Visa charge financial institutions, such as banks, for processing card payments on behalf of merchants are also set to increase, the report said.

[AdjustedEarnings]

I've got some basic business model questions about V/MA and hoping to hear your responses (on the ones you know the answers to). Thank you in advance.:

 

A. Are data processing revenues basically the 'fixed' component of each transaction and the service revenue the volume component of the same transaction? Or are they differentiated in any other way? On a related note, why does V report service revenues on a quarter lag?

 

B. In general, we all know V/MA get paid on GDV and # of transactions. Much of it is from merchants. But does anyone have ideas about how much issuers pay (either gross or net of client incentives)? Are their payments based on cards issued or transactions or volume? Or all three? (I understand it's all the same transaction so it's difficult so separate. I'm trying to use this information to determine the effect of banking consolidation on these players.)

 

C. How do V/MA share in the economics of co-brands?

 

D. What do V/MA get paid on transactions on their brand which are NOT processed by them? In such cases, who are some of the other networks processing these transactions? Are there any particular cases where this happens more? E.g. debit v/s credit, or intra-Europe, etc.

 

E. Where on the CF statement are capitalized client incentives? Because they cannot be seen there, my guess is that these are withheld by issuers in making their payments to V/MA. The revenue presentation is 'net'. So the capitalization in the CF statement is implied via the reduction in net-income v/s the way you'd capitalize anything else (e.g. PP&E). Can someone please confirm?

 

F. In Europe, is "Scheme" the same as brand? If so, is the Article 8 separation of Scheme and processing is the same as brand v/s processing?

 

Update: I do understand the cash-flows... meaning, how Visa simply keeps portion of the merchant discount, etc. The questions above are more about understanding "who pays who what and when" from a business/contractual standpoint rather than how the cash flows between parties as a matter of practice.

[SlowAppreciation]

I've got some basic business model questions about V/MA and hoping to hear your responses (on the ones you know the answers to). Thank you in advance.:

 

A. Are data processing revenues basically the 'fixed' component of each transaction and the service revenue the volume component of the same transaction? Or are they differentiated in any other way? On a related note, why does V report service revenues on a quarter lag?

 

B. In general, we all know V/MA get paid on GDV and # of transactions. Much of it is from merchants. But does anyone have ideas about how much issuers pay (either gross or net of client incentives)? Are their payments based on cards issued or transactions or volume? Or all three? (I understand it's all the same transaction so it's difficult so separate. I'm trying to use this information to determine the effect of banking consolidation on these players.)

 

C. How do V/MA share in the economics of co-brands?

 

D. What do V/MA get paid on transactions on their brand which are NOT processed by them? In such cases, who are some of the other networks processing these transactions? Are there any particular cases where this happens more? E.g. debit v/s credit, or intra-Europe, etc.

 

E. Where on the CF statement are capitalized client incentives? Because they cannot be seen there, my guess is that these are withheld by issuers in making their payments to V/MA. The revenue presentation is 'net'. So the capitalization in the CF statement is implied via the reduction in net-income v/s the way you'd capitalize anything else (e.g. PP&E). Can someone please confirm?

 

F. In Europe, is "Scheme" the same as brand? If so, is the Article 8 separation of Scheme and processing is the same as brand v/s processing?

 

Update: I do understand the cash-flows... meaning, how Visa simply keeps portion of the merchant discount, etc. The questions above are more about understanding "who pays who what and when" from a business/contractual standpoint rather than how the cash flows between parties as a matter of practice.

 

A. Yes, data processing fees are a flat fee per transaction, regardless of whether it's a $1 charge or a $1m charge. I think it's around $.073? Service fees are a function of payment volume. Historically this has been around .11% on the dollar amount of every transaction. All said, a $100 transaction will net Visa ~ $0.17.

 

B. Visa does not make money from merchants. It comes from banks. I think there are many variables that affect what they ultimately pay (e.g., transaction type, payment volume, fraud rates, chargebacks, etc). I would LOVE to see what one of these contracts actually looks like.

 

C. I'm not sure what the actual break down is and I'm sure it differs for each cobrand partner and bank. But I believe most of the time the issuer pays a "bounty" to the partner (e.g., airline or Costco) for every new sign up. Issuers benefit because they get a new sales channel of perhaps a highly desirable customer base (e.g., Costco customers, freq travelers, etc) who spend a lot on their cards (which means more interchange fees for them). Also probably some data sharing between parties. As far as benefits for V/MA... more and more frequent spend on cards is good. More acceptance at merchants is good. Cobrand cards help. (I'm sure there's more I'm missing though)

 

D. I believe I read somewhere they get paid the data processing fee, but not the service fee. Would love to know if that is true though.

 

F. I would think scheme = brand, but am not positive.

[cameronfen]

I think at some point, the fintech that has developed in emerging markets will realize they can target the United States.  Obviously card companies have a formidable network moat, but you are already seeing some chains in the US accept alipay (and why not the fees are like half the price as cards).  At the very least this will threaten pricing power. 

[mjm]

do not know much, but do know JPM entered contract whereby it's think its costs are fixed with V.

 

https://www.reuters.com/article/us-jpmorganchase-creditcards-insight/jpmorgan-uses-its-might-to-cut-costs-in-credit-card-market-idUSKCN0R80B620150908

[SlowAppreciation]

Buy-side analyst here who covers V and MA. Just want to clarify a couple points that have been made. In general I should note that V and MA are really bad at disclosing this stuff. It is indeed part of their secret sauce, so you can't really blame them, but it does make diligence pretty hard.

 

A. What's been said is roughly correct. Keep in mind that net revenue yields are much higher on credit than debit (and significantly higher than both when talking about cross-border).

 

B. V/MA charge both the issuers and merchant acquirers (e.g., Worldpay, Global Payments, etc.). But our work suggests that client incentives are disproportionally allocated towards the issuer side (makes sense given their concentration as you point out). Frankly V and MA are monopolies only on the merchant side; issuers have a choice of either V or MA and thus more leverage in extracting incentives.

 

C. Economics are generally no different than your average transaction, but given these customers are often more affluent and loyal than on average, V and MA compete aggressively here on incentives. Also, given co-branding relationships are often prevalent across an entire industry, individual players are unlikely to all use the same network because they don't want to share their merchant data (e.g., all of the major US airlines use different networks).

 

D. There's no hard-and-fast rule here, but we've heard it's roughly half (basically just service revenues / brand fees, and no transaction processing fees). There are a ton of debit networks out there, which has resulted in lower net revenue yields and processing share (e.g., STAR, NYCE, PULSE debit networks). Europe is the biggest place where V and MA often don't process, though there is not a pan-European network (many attempts have failed). Instead, various countries have domestic card networks (like Cartes Bancaires in France and Girocard in Germany). These are old, crappy networks that V and MA are trying to displace, but since they are still OWNED by the banks, V and MA haven't gained much momentum in winning share. That said, believe it or not many of these domestic networks can't process debit transactions online (e.g., Girocard), and since Europe is generally credit-averse, the rise of eComm might allow V and MA to make their value prop a bit more clear as time goes on.

 

E. This is exactly right. Sometimes incentives are reductions in cold hard cash due, and other times they are a reduction in fees after a certain amount of volume is reached. Best to track incentives / rebates as a % of gross revenues.

 

F. Yes, scheme = brand. The domestic networks I refer to above are often called "domestic schemes" in Europe, though they are really schemes / brands plus networks. (And then AMEX would be a scheme / brand plus network PLUS issuer.) This rule didn't cause much of a problem beyond requiring networks to break out these revenues separately. Note that some of the domestic networks in Europe like Girocard don't actually charge scheme fees, just interchange, which is regulated by the way at 30bps on credit and 20bps on debit in the EU.

 

Very helpful, thank you.

[AdjustedEarnings]

Excellent, thank you! What do you make of the recent cross border trends at V?

 

Buy-side analyst here who covers V and MA. Just want to clarify a couple points that have been made. In general I should note that V and MA are really bad at disclosing this stuff. It is indeed part of their secret sauce, so you can't really blame them, but it does make diligence pretty hard.

 

A. What's been said is roughly correct. Keep in mind that net revenue yields are much higher on credit than debit (and significantly higher than both when talking about cross-border).

 

B. V/MA charge both the issuers and merchant acquirers (e.g., Worldpay, Global Payments, etc.). But our work suggests that client incentives are disproportionally allocated towards the issuer side (makes sense given their concentration as you point out). Frankly V and MA are monopolies only on the merchant side; issuers have a choice of either V or MA and thus more leverage in extracting incentives.

 

C. Economics are generally no different than your average transaction, but given these customers are often more affluent and loyal than on average, V and MA compete aggressively here on incentives. Also, given co-branding relationships are often prevalent across an entire industry, individual players are unlikely to all use the same network because they don't want to share their merchant data (e.g., all of the major US airlines use different networks).

 

D. There's no hard-and-fast rule here, but we've heard it's roughly half (basically just service revenues / brand fees, and no transaction processing fees). There are a ton of debit networks out there, which has resulted in lower net revenue yields and processing share (e.g., STAR, NYCE, PULSE debit networks). Europe is the biggest place where V and MA often don't process, though there is not a pan-European network (many attempts have failed). Instead, various countries have domestic card networks (like Cartes Bancaires in France and Girocard in Germany). These are old, crappy networks that V and MA are trying to displace, but since they are still OWNED by the banks, V and MA haven't gained much momentum in winning share. That said, believe it or not many of these domestic networks can't process debit transactions online (e.g., Girocard), and since Europe is generally credit-averse, the rise of eComm might allow V and MA to make their value prop a bit more clear as time goes on.

 

E. This is exactly right. Sometimes incentives are reductions in cold hard cash due, and other times they are a reduction in fees after a certain amount of volume is reached. Best to track incentives / rebates as a % of gross revenues.

 

F. Yes, scheme = brand. The domestic networks I refer to above are often called "domestic schemes" in Europe, though they are really schemes / brands plus networks. (And then AMEX would be a scheme / brand plus network PLUS issuer.) This rule didn't cause much of a problem beyond requiring networks to break out these revenues separately. Note that some of the domestic networks in Europe like Girocard don't actually charge scheme fees, just interchange, which is regulated by the way at 30bps on credit and 20bps on debit in the EU.

[Liberty]

https://news.bloomberglaw.com/class-action/visa-mastercard-settlement-up-to-6-24b-gets-prelim-court-nod

[Jerry Capital]

Merchant Discount Rate = Interchange + Processing Fee + Network Fee

 

A typical transaction might look like the following:

 

Merchant Discount Rate = 200 basis points

 

Issuing Bank (Interchange or issuer fee) = 170 basis points

 

Merchant Bank (Merchant Acquirer Fee or processing fee) = 20 basis points

 

Network Fee (V/MA Fee) = 10 basis points

 

If this transaction was $100 the merchant would receive $98, the issuer say Royal Bank of Canada $1.7, V/MA 0.10$, the merchant bank say global payments/moneris would receive $0.20

 

Merchant Discount Rate = Net Interchange (Interchange - Cost of Rewards) + Processing Fee + Network Fee

 

Rewards = 100 basis points

 

You can make things a little more complicated adding in customer rewards which would be the same as above but the issuing bank has to fund the cost of rewards to get net interchange, on average its about 100 basis points (1% cash back, 1% in travel rewards). So the issuing bank would be left with 70 basis points. If the rewards are say 2% cash back then the MDR would simply go up from 2% to 3% (for example the MDR on a World Elite is something over 3%, which is why merchants don't like accepting these premium cards).

 

You could further explore the issuing banks profitability by taking net interchange and adding interest on late balances less the cost of bad credit.

 

Net Interchange +/- Net Interest or (Interchange - Cost of Rewards) +/- (Interest Revenue - Interest Expense)

 

The above numbers are extremely rough, and vary from country to country, I am sure people can poke holes in them but the magnitudes should be quite accurate for North America.

 

In terms of consolidation of the banking industry I would point out that some of the highest ROEs in banking come from the credit card issuing business. Also note where Apple took 10 to 15 basis points from when they introduced apple pay - the issuing banks...

 

If anyone wants to further add to the above equations it would be helpful, but for me I feel like the above is enough to generally understand the porters five forces in the industry.

[SlowAppreciation]

Does anyone know where I can find Visa's annual cross-border volumes (total $ amnt)? They disclose the growth each quarter and year, but I haven't been able to find the actual dollar amount. Or is this not shared and can only backed into from the estimated yield?

[SlowAppreciation]

Does anyone know where I can find Visa's annual cross-border volumes (total $ amnt)? They disclose the growth each quarter and year, but I haven't been able to find the actual dollar amount. Or is this not shared and can only backed into from the estimated yield?

 

Yep, not shared! It's huge though no doubt, perhaps close to 40% of net revenues

 

Bah that’s what I thought. Can’t it be backed into based on estimated yield though (around 1%?)

[AdjustedEarnings]

Where can I find reliable and up-to-date information on European interchange regulation? What does this regulation portend for regulation elsewhere, outside Europe? Could other countries go this far?

[AdjustedEarnings]

Not good for Visa Direct:

 

https://www.cnet.com/news/paypal-now-lets-you-instantly-transfer-money-to-bank-accounts/

 

The fact that PayPal has access to almost-RT ACH through JPM probably means Venmo will soon have this functionality too (currently I believe you need to do push payments using debit). Thankfully the fee is on par with network's push payments. Though, here, the chance of devolving into a price war are higher as their end customers (Venmo, Zelle, etc.) are not profitable and are still figuring out their monetization models. Let's hope commercial sense prevails.

[Liberty]

 

Visa CEO interview

[fuzzhead1506]

Current big picture thoughts on Visa:

 

5 year revenue growth of 13%

share reduction to 2100

Exit EV/Rev TTM multiple of 15.5x

Debt of 70B, Cash of 10B

--> Mkt cap of 539B

share price of $257 at year 5

plus dividends of $8 gets a $265 total return... that gets you between 10% and 11% returns from this level

 

Anything I am missing?

[SlowAppreciation]

Current big picture thoughts on Visa:

 

5 year revenue growth of 13%

share reduction to 2100

Exit EV/Rev TTM multiple of 15.5x

Debt of 70B, Cash of 10B

--> Mkt cap of 539B

share price of $257 at year 5

plus dividends of $8 gets a $265 total return... that gets you between 10% and 11% returns from this level

 

Anything I am missing?

 

You can look at it like this too:

[fuzzhead1506]

 

 

You can look at it like this too:

 

I suppose I lean bearish on my operating margins estimates.  I would think that if V gets that much margin expansion then we would see a rerate higher (not lower). Unless interest rates also went up rather drastically, that is. 

 

Do you have any thoughts on how that margin expansion might happen or were you just pulling numbers out of a hat, like me?  :P

[SlowAppreciation]

 

 

You can look at it like this too:

 

I suppose I lean bearish on my operating margins estimates.  I would think that if V gets that much margin expansion then we would see a rerate higher (not lower). Unless interest rates also went up rather drastically, that is. 

 

Do you have any thoughts on how that margin expansion might happen or were you just pulling numbers out of a hat, like me?  :P

 

Yeah it might be aggressive, though their business has tons of operating leverage so it’s not impossible. The incremental cost of processing each transaction is almost $0, and it costs the same to process a $.01 transaction as it does a $1,000 transaction. So Visa doesn’t need to grow expenses anywhere near the same rate as payment and transaction volume. As a result, operating margins should expand. Especially when they bring V Europe pricing to be more in line with the rest of the business.

[Jurgis]

Do you all incorporate possible regulatory pushback on fees? It's tough to estimate I grant that.

[SlowAppreciation]

Do you all incorporate possible regulatory pushback on fees? It's tough to estimate I grant that.

 

I'm not smart enough to estimate it, so I just bake in conservatism elsewhere. I figure cross-border fees are most at risk (either due to regulatory or market factors, or both), so I forecast its growth rate to be about half of what it has been for the past 5-10yrs. And some could argue that a 20x multiple is on the low side.

 

 

[TheAiGuy]

The multiples in those valuations are too probably low if you really get 13% revenue growth for five years....

 

Current big picture thoughts on Visa:

 

5 year revenue growth of 13%

share reduction to 2100

Exit EV/Rev TTM multiple of 15.5x

Debt of 70B, Cash of 10B

--> Mkt cap of 539B

share price of $257 at year 5

plus dividends of $8 gets a $265 total return... that gets you between 10% and 11% returns from this level

 

Anything I am missing?

 

You can look at it like this too:

[SlowAppreciation]

The multiples in those valuations are too probably low if you really get 13% revenue growth for five years....

 

Current big picture thoughts on Visa:

 

5 year revenue growth of 13%

share reduction to 2100

Exit EV/Rev TTM multiple of 15.5x

Debt of 70B, Cash of 10B

--> Mkt cap of 539B

share price of $257 at year 5

plus dividends of $8 gets a $265 total return... that gets you between 10% and 11% returns from this level

 

Anything I am missing?

 

You can look at it like this too:

 

Probably right.