Description

LONG: GMO Financial Gate

Summary

GMO Financial Gate (“GMO FG”) is fundamentally transforming Japan’s ¥300 trillion offline payments market by replacing legacy infrastructure to benefit the largest stakeholders in the offline cashless economy. The company is simultaneously an under-covered, cheaply-valued COVID-recovery beneficiary and a rapidly compounding tech disrupter. At an EBITDA-based exit multiple that is half of similarly growing global peers, we believe the stock would trade at ¥79,276/share in September 2024, offering nearly a 4x multiple on investment in 3.5 years.

The company’s new “Stera” terminal and recurring service was developed in partnership with the largest B2C credit card merchant acquirer in Japan, SMFG (25%+ share), and Visa’s VisaNet to replace and improve the country’s forty-year old duopoly switching networks operated by NTT Data and JCB.

Benefits to key stakeholders:

• SMFG – GMO FG eliminates ¥3+ of per-transaction fees, opens the large and previously inaccessible low-ticket vending machine market, and creates revenue-generating opportunities by improving transaction data transparency
  • Illustrative Savings – ¥45 billion/year based on SMFG’s projected credit card volumes if entirely migrated to Stera (¥3 * ¥30 trillion GTV / ¥2,000 average transaction size)
    
• Merchants – GMO FG lowers small merchants’ MDR by 40-50 bps (net of GMO FG’s take), improves the cashless receivables cycle (e.g., from biweekly to daily), reduces terminal replacement and fixed network costs, consolidates cashless payment contracts, provides simplified sales dashboards, and reduces the risk of chargebacks and data leakage
  • Illustrative Savings – ¥90,000/merchant/year (¥50 million annual sales * 40% credit card usage * 45bps)
    
• Consumers – GMO FG almost uniquely allows all relevant payment forms (contactless credit cards, QR, third-party and store-branded loyalty points, prepaid wallets, etc.) to be accepted by a single terminal
  
• Government – GMO FG accelerates the government’s effort to address retail labor shortages by improving Japan’s cashless penetration, which is close to the lowest among global peer economies

Description

GMO FG went live with its new Stera platform in July 2020. While the company’s current revenue mix is dominated by low margin terminal sales, we expect profits to be driven by spread income on gross transaction value (“GTV”), per-transaction fee income, recurring fixed fee income, and, in the medium-term, merchant and consumer financing services.

The total offline payments market in Japan is roughly ¥300 trillion, with an estimated 28% cashless penetration. This compares with Korea’s cashless penetration of over 90%, while Singapore and many other comparable markets are over 50%. The Japanese government has set a target of 40% cashless penetration by 2025 and 80% over the long-term, periodically providing incentives to consumers and merchants to increase the cashless ratio.

SMFG Visa credit cards account for 25%+ of the consumer credit card market in Japan; SMFG’s incentive to reduce per-transaction fees provides a strong starting point for GMO FG to gain scale. We believe GMO FG will eventually collect more than 10 bps on GTV and a blended ¥1/transaction on a large segment of Japan’s offline payments market. Our base case scenario puts the stock currently at ~7x 9/26E EBITDA, with only ~3% of the total offline payments market penetrated by then (i.e., high double digit EBITDA growth for many years beyond FY 9/26). GMO FG’s immediate parent company, GMO Payment Gateway (GMO PG), currently trades at 49x 2-year forward EBITDA with lower current forward growth than GMO FG would have in five years. We note Adyen and Square trade at even higher 2-year forward EBITDA multiples than GMO PG. We are assuming an exit EBITDA multiple of 25x for a target price of ¥79,276/share in September 2024.

Background

The company is part of the larger GMO Internet group, a leading technology conglomerate with interests in payments, data centers, security, e-signatures, and other growth markets in Japan. GMO Internet is led by Masatoshi Kumagai, who has received increasing attention since dining with Prime Minister Yoshihide Suga last October to discuss Japan’s digitization initiatives. While his approach of having many listed subsidiaries might raise concerns over conflicts of interest, Kumagai has repeatedly been able to align stakeholder and subsidiary management interests with those of minority shareholders in GMO group companies. Commenting on the president of GMO PG in a recent Bloomberg profile, Kumagai said, “Do you think he’d work that hard if he was merely the head of the payments division at GMO Internet? His motivation comes from being a president as well as a shareholder.”

GMO PG, which operates the online payments counterpart to GMO FG’s offline payments business, is the highest-valued group company so far. In addition to collecting an initial set-up fee and ongoing fixed fees from all of its merchant clients, GMO PG collects a ~50 bp spread on most of its merchants’ GTV (when the company acts as an aggregator though “Agent Contracts”), reducing otherwise high MDRs and complicated procedures for smaller online merchants. In recent years the company has pushed into “buy-now-pay-later” / “payment-after-delivery” services. This model has led to incredible compounding over the past 15 years (chart below), and more than a 100-fold return for shareholders in the past eleven years (160-fold in the twelve years since the financial crisis).

According to GMO PG, GMO FG is in the early stages of replicating GMO PG’s trajectory, substituting the tailwind of e-commerce penetration with offline cashless penetration; the offline cashless market has a significantly larger TAM than e-commerce.

While GMO FG had slowly been rolling out payment terminals to merchants prior to 2019, the company took a major step forward in February 2019 when it announced a joint agreement with SMFG and Visa, both longtime GMO PG partners, to develop a next-generation offline payment platform business. This was further cemented in October 2019 with a detailed announcement of the Stera platform. Prior to GMO FG’s IPO, GMO PG held ~65% of GMO FG, with the remainder held by GMO FG management and stakeholders including consumer finance companies and the venture arms of SMFG and other major banks. The IPO and launch of the Stera business were pushed back due to COVID, with both eventually coming in July 2020. GMO PG did not sell any shares in the IPO, and while GMO FG stock has performed well since then, the IPO came at the height of concerns over COVID’s impact on offline commerce and was accordingly priced low relative to the company’s long-term potential.

For nearly forty years NTT Data’s CAFIS has been the primary toll collector for credit card transactions in Japan; retail POS terminals have connected to the CAFIS system for security, authorization, switching, clearing, and packaged sales data, in exchange for a ¥3-5 per-transaction fee. Visa provides this service to most acquirers and merchants globally through its VisaNet service, but because of the continued high preference for cash in Japan, CAFIS has maintained its stranglehold on the market until now. The bypassing of CAFIS by Stera, via VisaNet, reduces cost, time, and overall friction for merchants, consumers, and payment companies.

Business Model

GMO FG is effectively a razor-and-blades business, and the company is in the early stage of growing its installed base of razors (in this case, Stera and other POS terminals/modules). There are roughly 4 million retail POS terminals in Japan with limited payment-form acceptance and an additional 4.5 million unattended “IoT” (vending, ticketing) machines with minimal cashless payment acceptance. As we mentioned above, merchants who already accept credit cards are incentivized to switch to GMO FG’s terminals for universal payment-form acceptance, including third-party and own-brand loyalty cards, simplified UI, and in many cases a lower MDR and a faster cash cycle. This is a major advantage over competing POS terminals, from Softbank’s PayPay for example, which do not accept competing e-money payment forms.

In its most recent quarterly briefing GMO FG improved disclosure of its terminal installed base to highlight the rapid growth of Stera terminals and IoT terminals.

Stera terminal installations increased QoQ from ~900 in the inaugural quarter to ~4,100 in the December quarter. 71.5% YoY overall terminal growth for an offline sales business during a pandemic demonstrates the strong value proposition and sales capability of GMO FG and its partners. The non-Stera terminals are also relevant since GMO FG will, in cases of Agent Contracts, negotiate a lower MDR and better terms for merchants on an aggregated basis, collecting a spread on GTV in addition to other recurring fees.

While most stakeholders are aligned with GMO FG, the clear loser is NTT Data’s CAFIS network. NTT Data has been anticipating the threat, which accelerated due to government initiatives to improve cashless payment options in recent years. In December 2018 NTT Data provided an update on its CAFIS business, highlighting its value to the payment ecosystem but indicating the potential for lower pricing. In 2020 CAFIS did effectively lower its pricing on transactions less than ¥1,000, targeting the IoT market which has until now effectively been precluded from credit card acceptance due to high per-transaction fees (on low transaction values). CAFIS will now charge ¥0.3 + 30 bps of GTV for transactions under ¥1,000.

We do not believe the transformation led by GMO FG is zero-sum. A more efficient payment platform driven by Stera will increase overall cashless penetration, including credit card usage at merchants who are currently cash-only. CAFIS will continue to collect per-transaction fees on the non-SMFG portion of credit card GTV. While GMO FG’s spread income might be earned at the expense of some credit card companies (lower MDRs), the increased GTV and lower merchant-related selling expense should more than make up the difference.

We note that SMFG has also partnered with Square, but this partnership is aimed at low-end merchants who can only afford an entry-level payment terminal. Square does not provide any cost savings to SMFG as Stera does, and we believe SMFG intends to upgrade these Square merchants to Stera eventually. One Japan banks analyst covering SMFG commented that Stera “is the cornerstone of SMFG’s cashless strategy.”

Why Now / Catalysts

We believe there are multiple well-known international brokers who are in the process of initiating coverage of the stock, as many of them already cover GMO PG. One broker recently initiated coverage of GMO FG, and we believe this analyst’s former employer is likely to initiate coverage soon.

The analyst who recently initiated has a FY 9/25 OP estimate that is 44% higher than the company’s target. When asked about his projections, the analyst said he believes he is being conservative on at least two assumptions: installed terminal base and recurring revenue gross margin. It is understandable that the analyst would like to be conservative; if he were applying more realistic assumptions, his FY 9/25 OP estimate would be several times the company’s own target, which would put his relationship with the company at risk. The company seems intent on under-promising and over-delivering.

Over the long-term the spread income on Agent Contracts and other GTV-driven financial services will be the primary bottom-line growth driver due to high incremental margins. In the near-term we believe spread income has a double tailwind; see the chart below on GMO FG’s Direct Contract and Agent Contract models:

The company only collects spread income on Agent Contracts, i.e., from merchants who have not been directly acquired by credit card companies and instead use GMO FG to negotiate better terms. In recent quarters approximately 30% of GMO FG’s GTV has been under Agent Contracts, and the company has collected roughly 30 bps of spread income on this volume (e.g., a small merchant’s MDR might be reduced from 3.25% to 2.5% + GMO FG’s 30 bps, for a net savings of 45 bps to the merchant). The company expects post-COVID economic reopening to disproportionately benefit the long-tail of small merchants, which are the same merchants more likely to be on Agent Contracts; i.e., the 30% ratio of Agent Contract GTV will go higher in the coming quarters (i.e., the 30% * 30 bps = 9 bps blended spread take rate will go higher). So in addition to the general offline commerce recovery tailwind, the company will benefit from the higher mix of Agent Contract GTV in the near-term.

Earnings Potential and Valuation

The company has provided some KPI targets for FY 9/25, in particular ¥5 trillion in GTV and ¥1.5 billion in OP. We make two observations on why these targets are conservative:

• The company has said its medium-term targets were made using a top-down estimate of 25% annual revenue growth and a 15% OP margin. The company has also said 25% annual growth is its target over the next fifteen years, and considering the early stage of penetration, we think 25% annual growth for the next four years is well below what it will actually achieve. In fact the company also explains the ¥5 trillion GMV estimate for FY 9/25 is based on gaining 1/3 of SMFG’s GTV; SMFG’s current GTV is already over ¥20 trillion (with some e-commerce) on a total offline cashless market of ~¥85 trillion, which in turn is on a total offline commerce market of ¥300 trillion. With overall cashless penetration moving to 40% in 9/25 (GMO FG and the government appear confident in this target), the total offline cashless market should be ~¥120 trillion, of which SMFG might have ~¥30 trillion (at constant market share); assuming some amount of double-counting for e-commerce, 1/3 of SMFG’s GTV would still result in ¥8-9 trillion of GTV for GMO FG. The company’s comment on the calculation above is that they want to stay prudent and will not explicitly discuss what is feasible, but they do not disagree with the assumptions.
  
• Since GMO PG has been a listed company for many years, we have data on how conservative the company has been with guidance and by extension how conservative subsidiary GMO FG might be with guidance. By early 2015 there was meaningful analyst coverage of GMO PG, and consensus FY 9/17 revenue estimates were ¥11.7 billion; 2.5 years later, the company’s actual FY 9/17 revenue was ¥21.1 billion, 80% higher than expectations.

Terminals typically have a five-year life, meaning by FY 9/25 nearly the entire current installed base of 4 million merchant terminals is addressable by GMO FG. The company estimates it could have an installed base of 300,000 Stera terminals by then, in addition to non-Stera terminals, unattended machine (“IoT”) terminals/modules (current addressable market of 4.5 million units), and partner terminals. We compare the company’s guidance with our FY 9/25 assumptions below:

• 100,000 Stera and 50,000 non-Stera terminal additions at a ¥65,000 ASP and a 12% gross margin
  • There is a one-time software customization component which results in the modestly positive margin, otherwise the company would be willing to sell at cost (this software margin is accepted by merchants)
    • The ¥65,000 price point is well below competing terminals, which cost over ¥100,000 with less functionality and fewer ongoing merchant benefits
  • 0% gross margin on IoT and partner terminals
    
• A blended average 12 bps of spread income on ¥8.1 trillion of GTV at a 100% gross margin
  • This includes the spread income on Active Contracts plus a small spread take rate on merchant financing (cash cycle, factoring)
  • The company is planning to provide more details on merchant financing and possibly on consumer financing in the next several quarters
    
• A ¥1 per-transaction fee on 3.2 billion transactions at a 70% gross margin
  • This is lower than analyst estimates because we believe the company will shift revenues more toward a GTV-based take-rate (in spread income) rather than a per-transaction fee, which is less feasible for small-ticket transactions
    
• A fixed fee of ~¥7,000/year on an average of ~560,000 terminals (including 261,000 Stera terminals), out of a total 8.5 million current addressable terminal market, which should expand with cashless penetration
  
• SG&A at 60% of gross profit based on a discount to long-term margin targets provided by the company
  • This is higher than GMO PG’s SG&A ratio even though selling activity for GMO FG is partially conducted by SMFG and GMO PG

We expect 33% top-line and 47% EBITDA growth from FY 9/25 to 9/26, at which point the company will still have only penetrated 5-10% of the offline cashless market and ~3% of the total offline payment market in Japan. We target an exit in 9/24 at ¥79,276/share, based on a two-year forward EBITDA multiple of 25x, which is a 50%+ discount to the average two-year forward EBITDA multiples of similarly fast-growing payment companies (Square, Adyen, and GMO PG) today.

Summarized Annual Financials

Comparable Valuations

Risks

Execution: GMO FG is a fast-growing company with a large market opportunity ahead, partnered with two of the largest companies in Japanese payments. We believe our estimates are reasonable, but they could be delayed by a few years if there are unforeseen bottlenecks in penetrating merchants.

Regulation: As mentioned above, GMO Internet Chairman Kumagai has personally discussed Japan’s digitization programs with Prime Minister Suga. The government’s effort to promote cashless payments is a major social undertaking, any change in administration priorities or pushback by representatives of tax-avoiding small merchants could reduce the company’s market opportunity.

Spread Income: According to our estimates, spread income would account for over half of gross profit in five years. We are assuming a blended low-teens bps take rate, including a few bps for basic merchant finance (with optionality on consumer finance). Merchant finance could be more competitive than we expect, and any competitor outside of GMO FG’s core business could try to take a share of GMO FG’s spread income if they can penetrate enough merchants with other services.

Catalyst

We believe there are multiple well-known international brokers who are in the process of initiating coverage of the stock, as many of them already cover GMO PG. One broker recently initiated coverage of GMO FG, and we believe this analyst’s former employer is likely to initiate coverage soon.

The analyst who recently initiated has a FY 9/25 OP estimate that is 44% higher than the company’s target. When asked about his projections, the analyst said he believes he is being conservative on at least two assumptions: installed terminal base and recurring revenue gross margin. It is understandable that the analyst would like to be conservative; if he were applying more realistic assumptions, his FY 9/25 OP estimate would be several times the company’s own target, which would put his relationship with the company at risk. The company seems intent on under-promising and over-delivering.

Over the long-term the spread income on Agent Contracts and other GTV-driven financial services will be the primary bottom-line growth driver due to high incremental margins. In the near-term we believe spread income has a double tailwind; see the chart below on GMO FG’s Direct Contract and Agent Contract models:

The company only collects spread income on Agent Contracts, i.e., from merchants who have not been directly acquired by credit card companies and instead use GMO FG to negotiate better terms. In recent quarters approximately 30% of GMO FG’s GTV has been under Agent Contracts, and the company has collected roughly 30 bps of spread income on this volume (e.g., a small merchant’s MDR might be reduced from 3.25% to 2.5% + GMO FG’s 30 bps, for a net savings of 45 bps to the merchant). The company expects post-COVID economic reopening to disproportionately benefit the long-tail of small merchants, which are the same merchants more likely to be on Agent Contracts; i.e., the 30% ratio of Agent Contract GTV will go higher in the coming quarters (i.e., the 30% * 30 bps = 9 bps blended spread take rate will go higher). So in addition to the general offline commerce recovery tailwind, the company will benefit from the higher mix of Agent Contract GTV in the near-term.