Haute Lumière

Commerce · III.04 · MMXXVI · daylight

La Bourse  /  Volume III  /  Nº III.04

Credit as a Commons

Volume III — Money, Energy, Information

Nine movements, one circle.

---

THE PLATE

---

THE LETTER

You have been taught that credit is a product — something a bank manufactures, prices, sells and regrets. That account is not wrong; it is merely the account of one supplier. It leaves out the older and much larger part of the system, which is that most credit in a working economy is extended sideways, between people who are not banks, and has always been.

Look at a set of company accounts. The line called trade payables is credit. Your supplier has lent you the goods and is waiting. So has theirs. Every firm in a supply chain is simultaneously a borrower and a lender, and the total of what they owe one another is, in most economies, larger than the stock of bank credit outstanding to the same firms. Nobody underwrote it. Nobody set a rate. It came into existence because trade came into existence, and it is held together by the ordinary expectation that people who deal with each other repeatedly will settle.

That is not a market. It is a commons.

And a commons has a carrying capacity. This is the part that neither the enthusiasts nor the regulators say clearly. A pasture can carry so many animals before the grass gives out; a credit commons can carry so much outstanding obligation before the promises stop clearing. The number is not fixed by sentiment or by solidarity. It is fixed by how much the members actually sell to one another, and how fast.

This chapter is about governing that carrying capacity rather than merely regulating the suppliers who operate inside it. It is a chapter with mathematics in it, because the central claim can be computed: a group of firms owing one another money can extinguish most of that debt without any money at all, and the share they can extinguish is a measurable property of who trades with whom.

It is also a chapter that names a threshold. Past a certain point, a mutual credit system stops being an alternative to banking and becomes an unusually badly capitalised bank. That point has a number, and the number is in here.

— The Editors

---

DISCOVERY

What is already working

The Swiss have been running one since 1934.

In the trough of the Depression, sixteen Swiss businesspeople — Werner Zimmermann and Paul Enz among them — founded the Wirtschaftsring, the economic circle, on a simple observation: their firms still had capacity, their customers still had needs, and what had vanished was not goods or demand but the franc in between. They opened a ledger. Members sold to one another and were credited in WIR, a unit worth one franc, spendable only inside the circle.

It began with roughly 300 firms. It has approximately 50,000 participating small and medium enterprises today — a factor of 166.7 — and it turned over CHF 1,430,000,000 in its own unit in 2013, which is CHF 28,600 per participant. It holds a Swiss banking licence. It has operated continuously for ninety years, through a world war, three franc shocks and two financial crises.

The finding that matters is James Stodder's. Studying WIR turnover against Swiss GDP across decades, he found the circle is countercyclical: WIR-denominated trade rises when the franc economy falls. When bank credit contracts, the circle expands, because that is precisely when a firm with stock and no cash prefers a customer who can pay in something. The Swiss have an automatic stabiliser sitting inside their small-business sector, and it was built by the small businesses.

Sardinia built one in a recession and it is still there.

In 2010 four young Sardinians with no finance background opened Sardex, a mutual credit circuit for island firms shut out of contracting bank lending. By the end of 2018 it carried about 3,200 company members and €43,000,000 of annual transaction volume — €13,437.50 per member. It is now the most closely studied mutual credit system in Europe, and the studies are unusually good: Paolo Dini and Laura Sartori got inside the brokerage operation and documented the thing that makes it work, which is not the software. It is that Sardex employs brokers who telephone members and introduce them to each other. The circle is hand-woven.

Whole countries clear obligations against one another as a public utility.

Since 1991 Slovenia has run a multilateral set-off of business debts as a national service, operated through its public records agency. Firms submit what they owe and are owed; the system finds the loops and cancels them; what remains is settled in cash. Tomaž Fleischman, Paolo Dini and Giuseppe Littera report that a typical round removes something in the order of 10 to 15 percent of participants' mutual obligations — with no money moving at all. In Kenya, Grassroots Economics reports that upwards of 70 percent of obligations inside the Sarafu network are clearable this way.

And the largest clearing house on earth does exactly the same arithmetic.

CLS settles over USD 8.0 trillion of foreign exchange payments a day across eighteen currencies. Its members fund about 1.0 percent of the value of their own instructions — under USD 10.0 billion per USD 1 trillion settled — because multilateral netting removes about 96.0 percent of the funding requirement. The most sophisticated settlement infrastructure in global finance and a slate behind a village bar are running the same algorithm at different scales.

The cooperative banks that came out of the crisis intact.

In Germany, Friedrich Wilhelm Raiffeisen (1818–1888) started a credit society at Flammersfeld in 1849 and the Heddesdorf lending association in 1864 on rules that now read like a design specification: one parish, unlimited joint liability, no dividends, an indivisible reserve, unpaid management. By 1914 Germany held over 34,000 registered cooperatives with 6,400,000 members — an average of 188.2 members each. The urban Schulze-Delitzsch credit societies numbered 945 with 620,000 members, or 656.1 apiece. Timothy Guinnane's archival work calls these societies information machines: their advantage was never cheap capital, it was that they knew things about a borrower no outside lender could learn at any price.

Their descendants are still here. Across Europe, cooperative banks came through 2008 with lower earnings volatility and lower loss rates than their listed comparators — a result documented by Rym Ayadi and colleagues at CEPS and by Wim Fonteyne at the IMF, both of whom went looking for the opposite.

Five systems, two centuries between them, one pattern: the credit was already there, held sideways between members, and what each institution added was not capital but a way of governing a shared stock.

---

THE ARITHMETIC

What clears, what does not, and where the line sits

First, the clearing itself, because it is the claim that sounds implausible.

Take eight firms in a valley — a tannery, a mill, a cutter, a maker, a shipper, a dyer, a printer and a bookkeeper — with sixteen obligations between them, from €6,000 to €52,000. Gross obligations: €339,000.

Try bilateral netting first, the obvious move. Pair by pair, cancel what each owes the other. In this circle it removes nothing at all, because no pair owes in both directions. Firms rarely do. Your supplier is not usually your customer.

Now net multilaterally. Compute each firm's receivables less its payables. The maker is owed €63,000 and owes €35,000: net +€28,000. The tannery owes €70,000 and is owed €49,000: net −€21,000. Do this for all eight and the positions sum to zero, because every euro owed is a euro owed to somebody.

The cash required to settle the whole system is the sum of the positive positions: €29,500.

  gross obligations                  €339,000
  settlement funding required         €29,500
  ------------------------------------------
  netting efficiency                    91.30 %
  cash released                      €309,500
  each euro of settlement clears        11.49  of trade

Ninety-one percent. And it can be pushed further, because a cycle of debt can be cancelled with no money whatsoever. Follow the loop tannery → mill → cutter → maker → shipper → tannery. The smallest obligation on it is €18,000. Subtract €18,000 from all five edges and every firm's position is unchanged — each has been relieved of €18,000 owed and €18,000 owing. That one cycle extinguishes €90,000, or 26.55 percent of gross, and nobody paid anything. Cancelling cycles greedily through this circle removes €272,000 — 80.24 percent — leaving an acyclic residue of €67,000.

State the denominator. This circle was constructed to contain cycles, so 91.30 percent is an upper bound and not a forecast. A real national population is sparser: Slovenia's monthly set-off clears 10 to 15 percent. CLS clears 96.0 percent because its members are all dealers in the same instruments. The spread between those two is 86.0 points, and the arithmetic is identical in every case. Netting efficiency is a property of how densely the obligation graph is looped — a fact about who trades with whom. It is never a fact about the software, and any promoter who quotes you an efficiency without quoting you a population is selling you a ledger.

Second, the carrying capacity.

A member's debit is only credit if the circle can clear it through trade. So measure it in the only unit that matters — months of the member's own in-network sales.

  clearing coverage ratio  =  debit balance / monthly in-network receipts

The tannery holds a €21,000 debit against €49,000 of annual in-network receipts, which is €4,083.33 a month. Its coverage ratio is 5.14 months. Against a policy cap of 3.0 months — a limit of €12,250.00 — it is €8,750.00 over, by 71.4 percent. Above that cap the balance has stopped being trade credit the circle can work off. It has become a loan, and the circle has no machinery to call one.

Third — the honest negative, and it is the chapter's cut.

Here is the identity nobody in the mutual credit literature says plainly. Take a circle of 220 members turning over €4,400,000 a year in-network, €20,000 each, with €640,000 of net debits outstanding — 14.55 percent of turnover, or 53.1 days.

The outstanding credits are also €640,000. They must be: the two sides of a mutual ledger sum to zero by construction. So the positive-balance holders are the depositors, the negative-balance holders are the borrowers, and between them sits an institution holding no capital at all against a €640,000 loan book.

A bank with that book at a 13.0 percent CET1 ratio would hold €83,200. The Basel III leverage floor alone would require €19,200. The circle holds nothing, which is a capital ratio of zero and a leverage ratio of infinity. In a 2.0 percent default year it loses €12,800, and because there is no capital, the loss falls as a 2.00 percent haircut on every positive balance in the system.

A mutual credit circle is not less levered than a bank. It is a bank read backwards, at infinite leverage, with no lender of last resort — and the only thing between it and a loss is governance. Which is why the right literature for it is Ostrom and not Basel.

The repair is arithmetic too, and it is small. A levy of 1.0 percent on in-network turnover raises €44,000 a year — a reserve ratio of 6.88 percent against the debits, enough to absorb a 6.88 percent default year. It clears the Basel leverage floor in 0.44 years and reaches a 13.0 percent CET1-equivalent in 1.89 years. One percent of turnover buys a commons a bank's capital position in under two years, and almost no mutual credit scheme charges it.

Fourth, the cooperative loss record, in both directions.

In 2024, US federally insured credit unions charged off 0.80 percent of average loans. FDIC-insured banks charged off 0.68 percent. The cooperative form lost more — 0.12 points more, a factor of 1.176.

Now the tail. Credit union net charge-offs peaked at 1.21 percent in 2009. Commercial bank net charge-offs peaked at 2.67 percent in 2010. The cooperative advantage in the crisis year is 1.46 points, a factor of 2.21.

Solve for the crisis frequency at which the two cancel:

  breakeven  =  0.12 / (1.46 + 0.12)  =  7.59 %   — one year in 13.17

Over a decade containing two crisis years the cooperative form is 1.96 points ahead cumulatively. Over a decade containing none, it loses. The cooperative discount is an insurance premium paid in good years, and an institution that sells it as a permanent efficiency will be embarrassed by the first quiet decade.

Fifth, joint liability, which is where the record is hardest.

Raiffeisen's unlimited joint liability worked in a village of perhaps a hundred households. It did not travel. Guinnane's study of the Irish transplant between 1894 and 1914 found societies that lent outside the parish, monitored nobody and sanctioned nothing — Ostrom's first, fourth and fifth principles failing at once — and they did not survive.

And when joint liability was finally tested experimentally, it did not earn its keep. Xavier Giné and Dean Karlan randomly removed group liability from existing Philippine lending groups and randomly assigned it in new areas. Over three years there was no increase in default where liability was removed, and the groups grew larger. In the expansion areas group liability produced no default advantage and fewer new groups formed: it deterred clients while delivering nothing. What did reduce default was meeting frequency — Benjamin Feigenberg, Erica Field and Rohini Pande found clients randomly assigned to weekly rather than monthly meetings were about 3.0 times less likely to default on the following loan. The mechanism was never the threat. It was the repetition.

The measured social cost sits alongside that. Fieldwork by Richard Montgomery and by Aminur Rahman documented peer seizure of household assets, coerced repayment and household conflict inside joint-liability groups — costs borne by borrowers and invisible in a lender's portfolio quality report.

Sixth, the founding story and the later evidence.

Muhammad Yunus's observation at Jobra in 1976 was that 42 people were held by USD 27.00 of debt — USD 0.6429 each. That observation was right, and it remains right: the binding constraint was the shape of the available capital.

The transformation claim is the part that did not hold. Six randomised evaluations published together in the American Economic Journal: Applied Economics in January 2015 — Bosnia, Ethiopia, India, Mexico, Mongolia, Morocco — found modest business investment and no transformative effect on consumption, income, health, schooling or women's empowerment. Rachael Meager's pooled analysis of seven of those experiments puts the average effect on household consumption near zero and estimates it precisely. In Hyderabad the take-up gap between treated and control areas was 8.8 percentage points, so scaling an intention-to-treat effect to the treated multiplies it by 11.36 — and an effect indistinguishable from zero, multiplied by eleven, is still indistinguishable from zero.

Hold that against a reported repayment rate of 97.0 percent and an implied loss rate of 3.0 percent, and the resolution is not a scandal. It is a definition. A repayment rate measures collection, not benefit. Suresh de Mel, David McKenzie and Christopher Woodruff measured real returns to capital in Sri Lankan microenterprises at 4.6 to 5.3 percent a month — 55.2 to 63.6 percent a year on the authors' own annualisation, 71.5 to 85.8 percent compounded — and found no positive return at all in enterprises owned by women. Against a 30.0 percent midpoint APR, the male-owned enterprises clear by 25.2 points and the female-owned ones are 30.0 points under water. That is a different country from Bangladesh and a different population from Grameen's, and it is stated here as a mechanism rather than a measurement — but the mechanism is the one that matters. A loan priced above the borrower's return is repaid out of something other than the investment.

---

DREAM

What becomes ordinary

In the economy where this is normal, a firm's treasurer opens two ledgers in the morning and thinks nothing of it.

The first is the bank account. The second is the circle: what the firm owes members, what members owe it, its debit line, and its clearing coverage ratio in months. The coverage ratio is the number the treasurer actually watches, the way a previous generation watched the overdraft. It is on the monthly pack. Everyone in the finance function can say what theirs is.

Settlement happens on the fifteenth. The circle publishes what it cleared, what it netted, and what cash was required — and the netting efficiency is a public figure, quoted the way a fund quotes its expense ratio, because members understand that it is the whole product. A circle running at 38 percent is a good circle. A circle that will not publish the number is not.

Trade credit is no longer an accident. A supplier extending sixty days knows exactly what that costs and knows that the obligation is clearable, because it sits in a graph with other obligations rather than alone in a ledger. Chains of firms that once all borrowed from the same bank to pay each other now cancel the loop and borrow for the residue. The bank is still there and is still needed; it is financing the acyclic part, which is the part that genuinely requires outside capital.

The reserve is boring and everyone knows the figure. One percent on turnover, into a fund that belongs to the members, published monthly against the outstanding debits. Nobody finds this radical. It is what an insurance mutual has always done, and a mutual credit circle without one looks, to this generation of treasurers, the way an uninsured warehouse looks now.

Credit unions and cooperative banks say the honest thing about themselves in public. Not we are safer — we are steadier, here is our loss rate in the last benign decade, here is our loss rate in the last crisis, and here is the crisis frequency at which the trade becomes worthwhile. They win members with that sentence, because it is the sentence a person who has been sold something before can believe.

And a village society somewhere is still doing it on a slate, with forty members and no software at all, at a netting efficiency the national system cannot touch, because everyone in it buys from everyone else.

---

DESIGN

The eight principles, applied to credit

Elinor Ostrom's contribution was never the discovery that commons work. It was eight design principles specifying the conditions under which they do not collapse, derived from irrigation systems, fisheries, forests and pastures that had held for centuries. Credit is a commons with the same shape, and the mapping is exact.

1 · Clearly defined boundaries. Who may hold a debit, and against whom. The credit union calls this the common bond; Raiffeisen called it the parish; Sardex calls it the island. Guinnane's Irish societies broke this first and everything else followed. The boundary is not exclusion for its own sake — it is what makes a default a fact somebody notices.

2 · Congruence between rules and local conditions. Debit lines set by the member's own in-network sales, not by a uniform limit. A firm that sells €4,083.33 a month inside the circle does not get the line of one selling ten times as much. The appropriation rule is proportional to the contribution, which is Ostrom's condition stated in a currency.

3 · Collective-choice arrangements. The members who carry the debits set the rules on the debits. One member, one vote — not one euro, one vote. This is the principle cooperative banks retain and demutualised ones abandon, and it is the best single predictor of which institutions behave cooperatively under stress.

4 · Monitoring, by monitors accountable to the members. Balances and coverage ratios visible to members, not only to staff. In a village this was free. Across 31,941.6 members per institution — the current US credit union average — it is not, and the honest thing to say is that statistical underwriting and a federal guarantee have replaced it.

5 · Graduated sanctions. Warning, then a frozen debit line, then suspension, then expulsion. Never straight to the extreme, and never nothing. Joint liability was a sanction with no gradations at all, which is why it was borne by the household rather than the borrower.

6 · Conflict-resolution mechanisms that are cheap and local. A members' panel that hears a dispute in a fortnight for nothing. The expense of the alternative is not the legal fee; it is that a member who cannot afford the process simply leaves, and takes their share of the circle's density with them.

7 · Minimal recognition of the right to organise. A legal form that holds — cooperative, mutual, credit union, licensed bank. WIR obtained a banking licence. Sardex operates as a company with a members' circuit. Systems that skip this spend their fourth year arguing with a tax authority about whether a clearing unit is income. It is; it is taxable at par; say so from the start.

8 · Nested enterprises. The circle inside a federation, the federation with a reserve, the reserve with an apex institution. This is the principle mutual credit schemes skip most often and the one that decides whether a circle survives its first bad year. A single circle with no reserve and no federation is a commons with no roof.

And the piece Ostrom does not supply, because her commons were physical: the capital rule. A pasture cannot be levered. A credit commons can. So the eight principles need a ninth term written in the constitution — the levy, its percentage, and the reserve ratio it is targeting — because the leverage identity above says that without it the governance is the only loss absorber, and governance is not a balance sheet.

---

DESTINY

How it holds when nobody is pushing

Three things keep a credit commons alive after its founders are tired, and they are the three that Ostrom's principles describe and that the surviving systems all have.

Density, deliberately manufactured. A circle sustains itself when members genuinely buy from one another, and that does not happen by registration. Sardex pays brokers to telephone members and introduce them. That cost is the product, not overhead — it is what raises netting efficiency, and netting efficiency is what pays for the circle. A scheme that treats matchmaking as a marketing expense will watch its efficiency decay towards the national average and its members towards the exit.

A reserve that is funded before it is needed. One percent of turnover, levied from the first transaction, published monthly. The temptation is always to waive it in year one to attract members, and the circles that do are the circles that discover in year four that their loss absorber is a conversation.

A published number that members check. Netting efficiency, coverage ratios, the reserve ratio. A commons is monitored or it is not governed, and monitoring by members costs nothing when the figures are simply put where members can see them.

Now the failure modes, named. It fails when the debit lines outrun in-network sales, which is the coverage ratio breaching quietly across many members at once rather than dramatically in one. It fails when a large member — the one everybody sells to — leaves, because the graph's density was resting on them and efficiency falls faster than membership does. It fails when the circle starts lending rather than clearing: the moment a debit is extended for a purpose other than a trade inside the circle, the balance has no route home. It fails when joint liability is bolted on as a substitute for a reserve, which transfers a balance-sheet problem onto households and buys, on the Philippine evidence, nothing at all in return. And it fails in the most ordinary way of all — the founders get older, the brokerage stops, and the circle becomes a directory.

---

DELIGHT

What it feels like

There is a specific pleasure in watching a clearing run finish. You submit what you owe and what you are owed; the cycles are found; and the number that comes back says that four fifths of it never needed money. It is not the pleasure of winning. It is the pleasure of discovering that a thing you had been carrying was mostly weightless.

Then there is the second pleasure, which lasts longer. You ring a member you have never met because the broker introduced you, and you find they make something you buy from three hundred miles away. The order is placed, and the money does not leave the valley. You notice, some months later, that you have started thinking of a dozen other firms as ours — not sentimentally, but in the way a person thinks about a shared road.

And there is the slate. Somebody still keeps one, and there is a reason it survived every technology that was supposed to replace it: a line drawn through a name in front of everyone is the cheapest monitoring system ever built, and it feels, to the person whose name it is, like being let back in.

---

OPERATIONALIZE THIS

At the level of finance

The instrument: a members' clearing circle with a turnover-linked debit line and a levy-funded reserve.

The structure. A cooperative or company limited by guarantee, owned by its members, operating a multilateral set-off ledger. A unit pegged 1:1 to the national currency. No interest on either side of a balance — this is what distinguishes a clearing circle from a lender and it is what keeps the tax and regulatory treatment simple.

The balance-sheet treatment. The circle's unit is not a liability of the operator. It is a bilateral trade obligation between members that the operator nets. Members carry balances as trade receivables and trade payables at par. Transactions are taxable at par in the national currency on the transaction date, exactly as a cash sale would be — settle this with your auditor and the tax authority in writing before the first trade, not in year four. The operator's own balance sheet carries only the reserve fund and its working capital.

The debit line. max debit = 3.0 × trailing three-month in-network sales, recomputed monthly, with a hard cap for new members. Published to the member. Breaching it freezes the line rather than suspending the member — a graduated sanction, per principle five.

The reserve. A levy of 1.0 percent on gross in-network turnover from the first transaction. On a circle of 220 members turning over €4,400,000 this raises €44,000 a year, clearing the Basel leverage floor equivalent in 0.44 years and a 13.0 percent CET1-equivalent in 1.89 years. Publish the reserve ratio monthly against outstanding debits.

The counterparty. A regional credit union or cooperative bank holds the reserve and provides the convertibility window. This is principle eight — nesting — and it is also the cheapest credibility available: a circle whose reserve sits at a licensed cooperative bank answers the only question a cautious member actually has.

The first ninety days.

The number that decides it. Not the member count. This one:

          gross cleared × netting efficiency × members' overdraft rate
          -----------------------------------------------------------   >  1
                     annual cost of running the circle

Worked: €4,400,000 cleared at 38.0 percent efficiency releases €1,672,000 of cash. At a 9.0 percent overdraft rate that is €150,480 of interest members no longer owe anyone. Against a €96,000 operating cost the ratio is 1.57×, and it clears.

Invert it and you get the figure to put on the front page of the board paper:

  breakeven netting efficiency  =  96,000 / (4,400,000 × 9.0%)  =  24.24 %

Below 24.24 percent this circle is a subsidy with a ledger attached. Above it, it funds itself out of interest its members no longer pay. Slovenia's national set-off runs at 10 to 15 percent; this circle needs 24.24; Sarafu reports 70. The margin here is 13.76 points, and the entire job of the brokerage is to defend it.

One business unit cannot carry this alone. A firm with €2,800,000 of annual trade payables across 140 suppliers, 22.0 percent of them inside the cluster, has €616,000 of clearable obligations; at 38.0 percent efficiency that releases €234,080 and saves €21,067 of interest — a ratio of 0.22× against the same €96,000 cost. It takes 4.56 firms of that size to fund the circle. That is the argument for joining one rather than building one, and it is the argument in a single number.

---

APPRECIATIVE QUESTIONS

Twelve, for a room

Discovery — what is already working

1. Which of our suppliers and customers are the same people, or one step apart — and when did that overlap last help us through a tight month?
2. Think of a time somebody here extended credit on judgement rather than on a score, and it was right. What did they know, and how did they know it?
3. Where in this region is trade already happening without money changing hands — barter, deferral, favours between firms — and who keeps track of it?

Dream — what becomes possible

1. If our trade payables and receivables cleared against our neighbours' every month, what would we do with the cash that stopped being necessary?
2. Imagine our monthly pack carries one number about the health of the credit we extend to each other. What is that number, and who would read it first?
3. If every member of this circle could see everyone's coverage ratio, what conversations would start that are not happening now?

Design — what we build

1. What is the smallest group of us whose obligations genuinely loop — and could we compute that this week from invoices we already have?
2. What would our graduated sanctions be, in order, and who would we trust to apply the first one?
3. If we levied one percent of turnover into a reserve from the first transaction, what would we want that fund to be allowed to do?

Destiny — how it holds

1. Who would do the introducing — the telephoning, the matchmaking — and how would we make that a paid role rather than a favour?
2. Which member is our density resting on, and what would we want in place before they ever left?
3. What is the first sign we would see that our debit lines have outrun our trade, and who would notice it first?

---

WORKS CITED

Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press.

Ostrom, E. (2005). Understanding Institutional Diversity. Princeton University Press.

Stodder, J. (2009). "Complementary Credit Networks and Macroeconomic Stability: Switzerland's Wirtschaftsring." Journal of Economic Behavior & Organization, 72(1), 79–95.

Stodder, J. and Lietaer, B. (2016). "The Macro-Stability of Swiss WIR-Bank Credits: Balance, Velocity and Leverage." Comparative Economic Studies, 58(4), 570–605.

Sartori, L. and Dini, P. (2016). "From Complementary Currency to Institution: A Micro-Macro Study of the Sardex Mutual Credit System." Stato e Mercato, 107, 273–304.

Littera, G., Sartori, L., Dini, P. and Antoniadis, P. (2017). "From an Idea to a Scalable Working Model: Merging Economic Benefits with Social Values in Sardex." International Journal of Community Currency Research, 21, 6–21.

Fleischman, T., Dini, P. and Littera, G. (2020). "Liquidity-Saving through Obligation-Clearing and Mutual Credit: An Effective Monetary Innovation for SMEs in Times of Crisis." Journal of Risk and Financial Management, 13(12), 295.

CLS Group. CLSSettlement and CLSNet public disclosures on netting efficiency and daily settlement values.

Guinnane, T. W. (1994). "A Failed Institutional Transplant: Raiffeisen's Credit Cooperatives in Ireland, 1894–1914." Explorations in Economic History, 31(1), 38–61.

Guinnane, T. W. (2001). "Cooperatives as Information Machines: German Rural Credit Cooperatives, 1883–1914." Journal of Economic History, 61(2), 366–389.

Fonteyne, W. (2007). Cooperative Banks in Europe — Policy Issues. IMF Working Paper WP/07/159.

Ayadi, R., Llewellyn, D. T., Schmidt, R. H., Arbak, E. and De Groen, W. P. (2010). Investigating Diversity in the Banking Sector in Europe. Centre for European Policy Studies.

National Credit Union Administration. Quarterly Credit Union Data Summary, successive quarters.

Federal Deposit Insurance Corporation. Quarterly Banking Profile, successive quarters.

Yunus, M. (1999). Banker to the Poor: Micro-Lending and the Battle Against World Poverty. PublicAffairs.

Banerjee, A., Karlan, D. and Zinman, J. (2015). "Six Randomized Evaluations of Microcredit: Introduction and Further Steps." American Economic Journal: Applied Economics, 7(1), 1–21.

Banerjee, A., Duflo, E., Glennerster, R. and Kinnan, C. (2015). "The Miracle of Microfinance? Evidence from a Randomized Evaluation." American Economic Journal: Applied Economics, 7(1), 22–53.

Crépon, B., Devoto, F., Duflo, E. and Parienté, W. (2015). "Estimating the Impact of Microcredit on Those Who Take It Up: Evidence from a Randomized Experiment in Morocco." American Economic Journal: Applied Economics, 7(1), 123–150.

Meager, R. (2019). "Understanding the Average Impact of Microcredit Expansions: A Bayesian Hierarchical Analysis of Seven Randomized Experiments." American Economic Journal: Applied Economics, 11(1), 57–91.

Giné, X. and Karlan, D. (2014). "Group versus Individual Liability: Short and Long Term Evidence from Philippine Microcredit Lending Groups." Journal of Development Economics, 107, 65–83.

Feigenberg, B., Field, E. and Pande, R. (2013). "The Economic Returns to Social Interaction: Experimental Evidence from Microfinance." Review of Economic Studies, 80(4), 1459–1483.

Montgomery, R. (1996). "Disciplining or Protecting the Poor? Avoiding the Social Costs of Peer Pressure in Micro-Credit Schemes." Journal of International Development, 8(2), 289–305.

Rahman, A. (1999). "Micro-Credit Initiatives for Equitable and Sustainable Development: Who Pays?" World Development, 27(1), 67–82.

de Mel, S., McKenzie, D. and Woodruff, C. (2008). "Returns to Capital in Microenterprises: Evidence from a Field Experiment." Quarterly Journal of Economics, 123(4), 1329–1372.

de Mel, S., McKenzie, D. and Woodruff, C. (2009). "Are Women More Credit Constrained? Experimental Evidence on Gender and Microenterprise Returns." American Economic Journal: Applied Economics, 1(3), 1–32.

Armendáriz, B. and Morduch, J. (2010). The Economics of Microfinance, 2nd edn. MIT Press.

Note on figures. The circle of eight, its netting efficiency, its cycle cancellation, the clearing coverage ratio, the leverage identity, the reserve arithmetic, the loss-rate breakeven, the monitoring scale ratios and the decision inequality are all computed in lib/verify/III_04.py and reproducible there, with their inputs, units and sources printed first. Where a published figure is a range it is carried through as a range. The circle of eight is declared as a constructed teaching case, and the real-world comparators are printed beside it for that reason.