Clearing the Hourglass

Dan Barabander wrote a great little article on what he calls “Hourglass Markets,” markets bottlenecked by bilateral interemediaries that bridge different trust layers, and the opportunity for crypto to disrupt them. The article starts with the problem of “stacking and funneling trust” in the banking system, or what Perry Mehrling would call the Hierarchy of Money.

Dan explains how the hierarchy of money in a single country (eg. we have deposits at the banks, who have deposits at the central bank) solves the problem of payments in that country (“stacking and funneling trust”). But the lack of an international central bank prevents the solution from being extended to the whole world, causing international payments to be an archetypal hourglass market requiring bilateral intermediaries to bridge the trust layers of different political jurisdictions.

The cool thing about the hierarchy of money is you technically only have to go up a level to the extent there is imbalance at a lower level. In other words, if you can do clearing at one level, you can limit your need to use currency from a higher level. Clearing is a multilateral function that can overcome the gross inefficiencies of bilateral relations. The largest banks for instance have CHIPS (Clearing House Interbank Payments System) which lets them clear their interbank debt before having to resort to moving bank reserves at the Fed. And its not just US banks, but US branches of foreign banks. So CHIPS is functionally a major international clearing house for interbank payments. But it’s still a tight closed club. And in 2001 they tightened it up heavily, changing the settlement system from running once end-of-day to running real-time netting with collateralized net positions. The new collateral requirements seem to have caused like 100 banks to drop out.

A big part of the problem of cross-border payments is that the post-WWII Bretton Woods settlement was incoherent. Where was everyone supposed to get the USD for settlement? And how inefficient and constraining it would be without international clearing. Of course there was an alternative proposal at Bretton Woods from the British, led by Keynes, for a multilateral clearing union (the Bancor), which would be the correct way to organize the international financial order across political regimes. This was actually originally EF Schumacher’s idea who was Keynes’ protege, but would later diverge from Keynsians heavily in economics. Alas, we did not get our precious Bancor, and got a USD cursed by Triffin’s dilemma instead.

In the aftermath of the war, without a clearing union, European countries formed a “spaghetti bowl” of hundreds of bilateral credit agreements. The worst sort of Hourglass Market. In 1950 they cut through the whole thing in one swoop by setting up a clearing house, the European Payments Union (EPU), which solved the problem outright and catalyzed the post-war boom.1 Apparently successful in its mission, in 1958 the EPU was tragically closed down. What a tragic end to an incredible international clearing institution. 1958 is also coincidentally the traditional birth year of the Eurodollar system.

In the absence of international multilateral clearing like the EPU, cross-border trade was again an Hourglass Market. What emerged to address it was the Eurodollar system, a dark web of countless bilateral USD credit relations between international banks. But this was also heavily supported by the launch of CHIPS in 1970. In the 1980s, 75% of the 140 CHIPS members were non-US banks. And this system seemingly worked for a while until its violent collapse in 2008. It never recovered.

Going further back in history, in the 1400 and 1500s, before the rise of central banks, the trust layers of different political sovereignties were bridged by a cabal of hundreds of international italian bankers who met every few months at payment fairs (in Geneva, and later in Lyons) to multilaterally clear virtually all european trade credit.2 In fact it was the existence of distinct political sovereignties that created an international FX market in the first place, even in this period, and enabled the international bankers to systematically profit from dealing in bills of exchange. Arguably, one of the major macro forces that killed the Medici for instance was the systematic imbalance in the payments graph between Flanders and Italy in the late 1400s that made it impossible to clear balances multilaterally without shipping gold.3 The system seemed to recover and peak in the 1500s at the fairs of Lyons (by this time the Medici were out of banking and into being popes and queens), but was destroyed by the Spanish, with their Genoese bankers, in the late 1500s using silver from the new world to mount an attack on the international clearing system by creating a new source of systematic imbalance: the permanent deficits of States.4

Coming back to modern day and FX, an international clearing house was finally set up in 2002 with CLS (Continuous Linked Settlement), which completely dominates. So its less clear to me that FX still has the Hourglass structure Dan describes, at least so long as the “hourglass” designation depends on international relations being bilateral (FX was a mess of bilateral settlement before CLS, which was the whole point of the clearing house). Perhaps there is still something to say about the Hourglass nature of markets even if they’re served by multilateral clearing houses, given the highly exclusive nature of those institutions. CLS only has 70 members. CHIPS 42. It’s not a dark web of bilateral relations per se, but the multilateral net is cast pretty tight. Of course members in both have complex webs of correspondent relations to extend indirect access to thousands of others (fun aside, the nostro/vostro accounts they use for this date back to the revolution in 14th c banking).

The question I’m asking is how does this all change if we introduce a multilateral clearing system accessible to all? Not an exclusive club for the largest institutions brokering access for everyone else, but a bonafide open clearing protocol that anyone can participate in. What if clearing, the most powerful financial capability in history, could be accessed by anyone?

This is the Cycles opportunity. Lets smash the hourglass.


  1. See “The European Payments Union” in Amato and Fantacci’s excellent book, “The End of Finance,” which puts the problem of clearing into its full historical perspective. ↩︎

  2. See the very important and understudied book, “Private Money and Public Currencies: the 16th century challenge”, by Boyer-Xambeau et al. It’s unfortunately a quite difficult book. I hosted a reading group for it at the Mimbres School which you can find there. ↩︎

  3. See De Roover’s “The Rise and Decline of the Medici Bank: 1397-1494” ↩︎

  4. Addressing the problems engendered by the permanent deficits of states was supposed to be one of the raisons d’etre of the crypto industry. Alas, I often find myself asking: https://www.youtube.com/watch?v=Pb8_Tu9Os-U ↩︎

comments powered by Disqus