Two-sided retail-payment networks such as Alipay, Mastercard and Visa enjoy the limelight, enable commerce worldwide, and are taken for granted by billions of consumers and tens of millions of merchants.

Near-invisible payment “hubs” are also two-sided networks, and while not glamorous, are increasingly indispensable for managing and delivering payments.

At their most limited, hubs within a single institution — typically a bank, funnel payments from multiple channels such as branches, ATMs and online, enable sharing common functions and/or provide a single intelligent gateway to external networks and firms. At their broadest, they facilitate any institution paying any other institution by any means. More often than not, however, they’ve been solving pedestrian internal operational issues.

Mastercard has made several moves to support the development of payment hubs.

In most of the developed world the lion’s share of electronic B-to-B, B-to-C, P-to-P and, excepting retail C2B, payments are from bank Demand Deposit Account to Demand Deposit Account (DDA to DDA), the economy’s anchor liquidity instrument.

While hubs may support consumer payments operationally, they aren’t retail systems. For retail payment systems powerful brands conveying the promise to consumers, merchants and financial institutions are essential. For spontaneous transactions between parties unknown to each other, they provide needed certainty. In contrast, hub brand requirements are comparatively modest. They communicate reputation and trust to a relatively small set of sophisticated firms that they hope to serve.

Retail-payment systems must build critical mass on both sides of the network before delivering value. For hubs the initial effort may be less daunting, but nonetheless, there’s an upfront cost to credential and securely connect to new payers and payees.

While retail networks’ and hubs’ marginal payments costs are both zero, their fees differ. The most powerful retail systems earn licensing and processing fees. Hubs routing payments typically collect just processing or switch fees. With scale both business models are appealing as core platform costs are substantially fixed.

A range of software providers, payment processors and networks directly and indirectly serve the hub market.

Payments mainstays believe hubs matter. Visionary entrepreneur Michel Akkerman’s Clear2Pay’s forté was SOA-compliant payment-hub software for banks. The processor FIS acquired Clear2Pay to add a state-of-the-art hub to its bank product suite.

The payments-software gorilla ACI Worldwide and Finastra license on-premise and hosted hub software, principally to banks for internal use. ACI Worldwide’s licensed software touches electronic payments at each stage in the value chain. In retail payments it crossed the processing Rubicon by acquiring the processor Payon, which is a global white-label gateway between merchant acquirers and payment networks. It wouldn’t be a big step for it to process payments between banks, processors and networks using its software, and beyond.

New and nontraditional entrants are taking a more expansive view of delivering payments interconnectivity without the enterprise.

Earthport’s central hub-and-spoke model cleverly connects domestic payment schemes with a network of nostro accounts (typically bank’s money held by another) worldwide. The plucky Modo Payments has a universal vision, aiming to be the Switzerland of global payment hubs, and launched with Klarna, Bank of America, Verifone and FIS. And through a program helping young, innovative payments companies achieve liftoff velocity, Mastercard is making introductions to its licensees.

Mastercard’s core franchise and DNA are retail, but it may be in the pole position to provide real-time national interbank delivery systems and stitch them together planetwide. It acquired U.K. banks’ interbank payments utility Vocalink. Notwithstanding its U.K. focus, Vocalink provided faster-payments technology in Sweden, Singapore and to U.S. banks’ ACH processor the Clearing House. Mastercard, however, has global reach. It could credibly introduce the platform to any market on earth, as nodes on a new Mastercard network.

Banks still control key payments infrastructure. For more than four decades Swift’s payment messaging system with a web of correspondent banks has enjoyed a near-monopoly on cross-border fund transfers between 11,000 banks worldwide. However, noncommercial monopolists exposed to competition have to be counted at risk.

Every country has at least one domestic interbank payments network and processor. They do what they were built to do, providing reliable, low-cost, nonspontaneous payments, generally between known parties. They haven’t, however, been enterprising.

Real-time interoperability between national ACH systems is intuitively appealing and a globalist’s dream. A few domestic players have made international moves. Italian banks’ payment utility SIA will support a pan-EU instant payments scheme. Vocalink has beachheads abroad. Worldline Equens processes interbank payments in the Netherlands and Germany. And French banks’ STETS processes ACH in France and Belgium.

In the U.S., the bank association NACHA manages ACH system rules and the Clearing House and Atlanta Fed are the primary processors. NACHA is thinking bifocally, improving the quality and speed of domestic ACH, while socializing ISO 20022 standards to better position itself to support cross-border payments. It’s ratcheted up domestic clearing and settlement speeds. As ISO 20022 is adopted, in theory, it’ll become easier for national systems to interoperate. Still a web of bilateral processing relationships among several hundred national ACH systems is not obviously the best or most efficient solution.

A handful of central hubs competing worldwide would be more dynamic, efficient and adaptive to changing market conditions and needs.

It would be folly to bank on universal standards. The global payments patchwork will be messy for the foreseeable future. In The Future and Its Enemies former Reason magazine editor Virginia Postrel makes a compelling case that seemingly chaotic competition and experimental innovation outperform centrally engineered and imposed solutions. Systems that can deal with change, complexity and system, national, and regional variations are more likely to prosper and deliver sustainable value than those relying on conformance to central or intergovernmental planning.

There are revolutionary alternatives. Bitcoin’s been hyped as a seismic disruptor of retail payments, money transfer and money writ large. It’s a new and specialized means of exchange, a speculative instrument, a poor unit of account because of its volatility, and, as a payment system lacks critical mass, a compelling licit use case, and reasonable cost and performance. While one could imagine bitcoin as an unmanaged and global electronic payment hub in the commons, unmanaged it’s not fit for purpose.

Would-be disruptor Ripple is pursuing a multipronged strategy, licensing software and a ledger gateway to banks to better manage cross-border payments, and curating its own digital currency XRP, which, ultimately, could be a (the) media of international value exchange.

There are enormous hub niches such as cross-border money transfer between MNOs and MTOs. Mastercard’s, eservGlobal’s and BICS’s joint venture Homesend, run by Mastercard (Orbiscom) alumnus Stephen Doyle, is a specialist addressing this high-growth market. Mauritius-based hub MFS also focuses on money transfer in emerging markets. It connects 120 million mobile wallets served by sub-Saharan African MNOs.

A competitive patchwork is evolving. Entrants from adjacent space like Mastercard and ACI Worldwide, and upstarts like Earthport, Modo and Ripple could expand the market and upend complacent incumbents.

Competition might be at the relationship level and/or at the transaction level with promiscuous partnering and transactions situationally routed based on hubs’ reach, approach and economics. Several of these hub approaches may be more appealing than others to industry payments cognoscenti, but it’s best to let the rough and tumble of the market pick winners and losers.