The Ponzi That Sank a Country

Part 1Half a country, all in: the scale of Albania's 1997 pyramid schemes

In the autumn of 1996, Albania's largest financial institutions were not banks. They were a holding company that ran supermarkets and package tours, an investment fund in the port of Vlorë and a "charity foundation" that collected cash in the villages around Lushnjë and Fier. When they stopped paying, eight of them owed 177.1 billion lek to 1,687,000 creditors, according to the reconstruction by Christopher Jarvis, an IMF economist, from administrators' and auditors' reports and Bank of Albania data. At the end-1996 exchange rate of 103.7 lek to the dollar, that is $1.7 billion: 64% of the country's GDP that year, owed to creditors equal to 51% of its people (Figure 1).

Set that against the formal system and the hierarchy is plain. At the end of 1996, Albania's entire broad money, cash in circulation plus every lek and dollar deposited in its banks, came to 154.6 billion lek. The schemes owed 1.15 times that, by Le Détroit's calculation. They were not on the fringe of the financial system. For the people who used them, they were the financial system.

The headline numbers vary, and the variation is itself instructive. Writing in the IMF's Finance & Development in 2000, Jarvis put the face value of the schemes' liabilities at $1.2 billion in November 1996, "almost half" of GDP, and said about two-thirds of Albanians invested. His own table gives 64% and 51%. The figures measure different things. The $1.2 billion is the face value in November 1996, about 45% of GDP by our arithmetic; the $1.7 billion is what the schemes owed their creditors when they collapsed, estimated later from administrators' and auditors' reports, after more weeks of deposits and of interest promised at 8 to 47% a month. Jarvis does not reconcile his two population figures; in our reading, "two-thirds" counts people in households that invested, while 51% counts the creditors on the schemes' books. Contemporary reports were looser still: The Christian Science Monitor put the sums invested over five years at about $3 billion on 6 March 1997, and the losses at about $1 billion two weeks later. No source disputes the order of magnitude: roughly half a country, and roughly half an economy.

The collapse took the state down with it. Protests began in January 1997 and became, in the south, an armed revolt. Arms depots were emptied, a government fell and an Italian-led force of about 7,000 troops arrived under a UN mandate. Estimates of the dead range from about 1,600, the figure given by the Inter-Parliamentary Union, to some 2,000 in Jarvis's work and over 3,000 in Albanian press accounts. This is how it happened, what it looked like from inside a bank, and why its warning signs matter again.

Part 2A financial system with no front door

Albania in the 1990s did not lack money. It lacked a safe place to put it. When the communist system ended in 1991, its state banks could not meet the demand for credit, Jarvis writes, and an informal market grew in their place, built on family ties and financed by remittances from Albanians working abroad. Those transfers were worth about $300 million a year, some 12% of GDP. They arrived as cash, in drachmas, lire, marks and dollars, as the anthropologist Smoki Musaraj describes in Tales from Albarado, her study of the period.

The price gap between the two systems was enormous. In 1992, a one-year deposit at a formal bank paid 32% a year; informal lenders paid an average of 240%, according to figures from ACER (1996) that the economist Dirk Bezemer published in 1999. By 1995 the gap had narrowed but not closed: 13.5% against 72%. In that period, Bezemer notes, the deposit-taking "companies" offered 8 to 12% a month for six-month money, and the "foundations" 50 to 100% for three or six months. Some of the early returns may have been real. Jarvis records the belief that smuggling in breach of the UN sanctions on Yugoslavia financed the high returns of some large companies; the sanctions were suspended at the end of 1995, and that income went with them.

The law was not silent. Law No. 8075 of 22 February 1996 provided that no one other than a bank could accept deposits from households, Jarvis notes. Nobody enforced it against the schemes. Bezemer described the resulting regime as "very strict for official banks but extremely lenient for informal financial intermediaries". The front door of the financial system was narrow, guarded and paid little. The side door was wide open, paid a fortune and was, for most people, the only one they knew.

That is Le Détroit's first reading: the schemes grew not because Albanians were uniquely credulous, but because a country with savings and no functioning banks will build its own intermediaries. Musaraj places the firms "at the interstices of formal and informal economic institutions". The danger lay in the gap between the two, and in the fact that no authority regarded the gap as its job.

Part 3Companies and "charity foundations": the look of a real business

Two different machines carried the same label. A World Bank typology from 1997, reproduced by Bezemer, sorted the main schemes into "companies" (VEFA, Kamberi, Gjallica), which had operated for three to five years, took unlimited deposits and accepted foreign currency, and "charity foundations" (Xhaferri, spelled Xhafferi in the IMF tables, Populli and Sude), which took lek only; Xhaferri capped deposits at 3,000 lek and Populli at 10,000.

The split mattered more than the label. By Le Détroit's arithmetic on Jarvis's table, the three foundations held 89% of all creditors but only 23% of the money. The average Xhafferi creditor was owed $250 and the average Populli creditor $200. The companies held 77% of the money with far fewer creditors: VEFA's average creditor was owed $7,707, Gjallica's $6,053. With the average wage at about $80 a month, as RFE/RL reported in January 1997, a typical VEFA creditor had placed about eight years of earnings; a typical Xhafferi creditor, about three months. The foundations were a mass lottery for the rural poor. The companies were where families with remittances and property put their capital.

The companies also looked like businesses, because in part they were. VEFA was founded as a commercial company in August 1992 and ran more than 200 activities in Albania, with joint ventures, and agencies in at least eight other countries, according to a European Court of Human Rights decision. It owned supermarkets and petrol stations, the journalist Gavin Haynes writes. The two advertisements below show the shopfront: Kamberi offering tourism at the Llogara pass, pig and poultry farming and sport; VEFA selling six-day coach trips to Budapest for 280 dollars from an office in Tirana's Palace of Culture. Neither mentions an interest rate. A 1996 VEFA advertisement reproduced with Musaraj's research celebrated the firm's fifth anniversary as the "incarnation of humane capitalism".

Shopfronts: how two schemes advertised themselves in 1996

Not a word about interest rates: the companies sold tourism, food and travel

Kamberi Company, Vlorë. A trading, tourism and farming group, by its own description. Kamberi promised 10% a month by July 1996.

The KAMBERI company, with its trading and production network, is present in several cities of Albania, especially in the south.Kompania "KAMBERI" me rrjetin e saj tregtar e prodhues është e pranishme në disa qytete të Shqipërisë, veçanërisht në ato të Jugut.Translated by Le DétroitKey line, transcribed as printed; ellipses mark omitted text

It is open to partnership in tourism, pig and poultry farming, sport and other activities. uncertain readingAjo është e hapur për bashkëpunim në këto aktivitete: turizëm, rritje dhe mbarështim derrash dhe pulash, sport ...Translated by Le DétroitThe words for pigs and chickens are partly blurred in the scan

... fresh mountain air with warm, iodine-rich sea air ... Llogara (a tourist complex on the pass above Vlorë). Everything fresh! uncertain reading... ajrit të freskët malor me atë të jodizuar e të ngrohtë detar ... Llogara. Gjithçka e freskët!Translated by Le DétroitThe lines printed over the photograph are partly illegible

Address: "Rakip Malili" street, building 1336, Vlorë uncertain readingAdresa: Rr. "Rakip Malili", Pall. 1336, VlorëTranslated by Le DétroitStreet name, building number and the telephone and fax numbers are hard to read

VEFA Holding, Tirana. The largest scheme by liabilities sold package trips from the Palace of Culture on Skanderbeg Square.

VEFA HOLDING offers you the pleasure of tourist trips abroad WITH VEFA TURIST: fantastic trips, reasonable pricesVEFA HOLDING ju ofron kënaqësinë e udhëtimeve turistike jashtë shtetit ME VEFA TURIST: udhëtime fantastike, çmime të arsyeshmeTranslated by Le DétroitKey line, transcribed as printed; ellipses mark omitted text

... six-day trips to Budapest (Hungary): five nights in Budapest and Yugoslavia; visas, meals, transport and guides included... udhëtimeve gjashtë ditore në Budapest (Hungari) ... pesë netë fjetje në Budapest e JugosllaviTranslated by Le DétroitThe agency also arranges Hungarian and Yugoslav visas, rooms, breakfasts and dinners

Price only 280 USD. Bookings every day until 27 November 1996 (new passport and 5 photographs required)Pagesa vetëm 280 USD. Prenotimet ... deri më dt. 27 nëntor 1996Translated by Le Détroit27 November 1996 is the booking deadline, not a departure date; the print has the typo "Frenotimet" for "Prenotimet" (bookings)

VEFA Travel Agency, Palace of Culture, Skanderbeg Square, TiranaAGJENSIA E UDHETIMEVE TURISTIKE "VEFA", PALLATI I KULTURES, SHESHI "SKENDERBEJ", TIRANETranslated by Le DétroitKey line, transcribed as printed; ellipses mark omitted text

The shopfront was real; it was simply far too small for the promise. Administrators later estimated VEFA's remaining assets at 7% of what it owed and Gjallica's at 1%. Jarvis drew the line between companies with real investments and "pure" pyramids. The line made no difference to the outcome: both kinds paid old depositors with new money. It made a large difference to how they were perceived, and the misleading distinction would return, in an official warning, at the worst possible moment.

Part 4The arithmetic of a Ponzi scheme

A pyramid does not fail when it runs out of money. It is insolvent from its first month; it fails when it runs out of new money. Jarvis's working paper makes the point with a hypothetical scheme, rebuilt in Figure 2. It promises 10% a month, invests half of each deposit and keeps half as cash; depositors redeposit part of their interest. In January it takes 100 and pays nothing, yet its net worth is already minus 30, because the promised interest is a debt the moment it is promised.

From there the books follow a fixed path. Cash flow stays positive while deposits grow, peaks in April and turns negative in June, when withdrawals and interest overtake new money. The operator then does the only thing that works: he raises the rate, to 20% in July, 30% in September and 40% in November. Each rise brings one good month of fresh deposits and a bigger debt. By December no one new arrives, the cash runs out after 686 of withdrawals, and the scheme owes 10,304 against assets of 1,287: twelve units for every hundred owed.

What Jarvis's figure shows next is who pays. A sliver of early investors doubled their money or more. Many made smaller profits, many lost part of their stake, and a large group, the late arrivals, lost everything. His published paper summed up the real outcome in Albania in a single sentence.

A pyramid destroys little wealth in aggregate. It moves it, from the last in to the first in and to the operators.

The real schemes traced the same path. The leading companies paid 4 to 5% a month in 1995, raised their rates to 6% in January 1996 and to 8% before the May elections; Kamberi went to 10% in July; Populli offered more than 30% in September; in November Xhafferi offered to treble deposits in three months and Sude to double them in two, Jarvis records. That is Le Détroit's second reading. The signal to watch in a scheme is not the level of its rate but its direction: a rising rate is a scheme borrowing at a higher price to repay old debts.

Part 5Rates that could not be real

The most dangerous rate in the story is the most modest-looking one. Next to Xhafferi's 47% a month, VEFA's 8% sounded almost prudent. Compounded, it is 152% a year: money doubles in nine months (Figure 3). After 1996 inflation of 12.7%, that is a real return of about 123% a year. Ten per cent a month, the rate of Gjallica, Kamberi, Silva and Cenaj, is 214% a year. At Sude's 41%, money doubles in two months; at Xhafferi's 47%, 100 lek becomes about 10,200 in a year.

Against that, the formal system offered almost nothing. The Bank of Albania raised the minimum rate on one-year deposits from 13.25% to 16% in 1996, Jarvis notes; at 16%, savings take more than four and a half years to double. Even the informal lenders, at 72% a year in 1995, look tame beside the schemes. Jarvis's own chart of the informal market shows monthly deposit rates of 5 to 6.5% early in 1996, about 8% from May and a peak above 10% in November: by then the "market rate" and the schemes' rates were the same thing.

No legitimate business could have paid these sums out of profits. To pay 8% a month honestly, VEFA's supermarkets, tours and petrol stations would have had to earn more than 150% a year on every lek entrusted to them, and to keep doing so as deposits multiplied. The higher-rate foundations did not even pretend. Yet the believable rate did the most damage: VEFA, at 8%, ended up owing $655 million, more than any other scheme. That is why we chose its rate for our title. A promise that sounds sensible is the one that attracts serious money, and serious money is what turns a fraud into a national crisis.

Part 6The boom that hid the bubble

Albania's macroeconomic figures were at their best exactly while the schemes grew, and that was no coincidence. After output fell by 27.7% in 1991, the economy grew by 9.6%, 9.4%, 8.9% and 9.1% in the four years from 1993 to 1996, according to the series Bezemer compiled from the OECD, EBRD, World Bank, IMF and Albanian sources. Inflation fell from 226% in 1992 to 7.8% in 1995. The lek held between 94.5 and 103.7 to the dollar. Unemployment fell from 29% to 12.4% (Figure 4). By the usual indicators, Albania was the transition's star pupil.

The same table shows what the indicators concealed. The budget deficit was still 11.3% of GDP in 1996. Imports were four times exports, with a trade deficit of $692 million. The economy was running on remittances, aid and imports, and its savings were flowing into institutions that no statistic captured. Bezemer called this the "Albanian paradox" and read it as a case of Hyman Minsky's financial instability hypothesis. His warning is the one that matters for analysts today: identifying the trend "is hindered if policy makers and observers largely rely on macroeconomic indicators".

Then the paradox resolved itself. In 1997 output fell by 7%, the lek lost almost a third of its value to end the year at 149.8 to the dollar, and average inflation rose to 32.1%. Jarvis put inflation above 40%, a figure that measures prices at the end of the period rather than the yearly average. He also judged that the schemes' direct economic effects were "difficult to specify and appear to have been limited": much of the money had changed hands rather than vanished. Le Détroit reads the episode differently on one point. The damage was not in the national accounts. It was in the state, which lost control of its territory, its arms depots and its credibility in the space of three months.

Part 7Warnings on paper: the IMF, the World Bank and the Meksi letter

Everyone who needed to know, knew. The IMF and the World Bank had expressed concern as early as December 1994 about the dangers of criminal enterprises, a worry as much about money laundering as about pyramid schemes, Jarvis writes. The strong warning came only in August 1996, when the Bank of Albania raised its concerns with a visiting IMF mission, which left a letter with the president urging early action. That month, on 20 August 1996, Prime Minister Aleksandër Meksi sent a secret letter, No. 2858, to the finance minister, Ridvan Bode, and the Bank of Albania governor, Kristaq Luniku, copied to President Sali Berisha. It described "extremely bad developments" in high-interest lending by VEFA, Gjallica, Kamberi and Sude and by the newer Silva and Populli, warned of economic, social, political and criminal risks, and asked for tax monitoring, compulsory balance sheets, foreign experts and the immediate closure of the newer firms, according to the journalist Hamdi Jupe, who published the letter in Shqiptarja in 2021 (translated by Le Détroit).

None of that happened in time. The schemes were close to power. Several companies contributed to the governing Democratic Party's campaign for the May 1996 elections; senior officials attended their events; and in November 1996 the prime minister and the speaker of parliament accepted medals at one company's anniversary celebrations, Jarvis records. Ministers told television audiences that the schemes were safe, Haynes writes. A committee set up in November 1996 to deal with the problem never met.

The public warning, when it came, made things worse. The finance ministry spoke out only in October 1996, and drew what Jarvis calls a misleading distinction between companies, implying that some schemes were sound. Parliament passed a law in November creating a panel to supervise them, with little visible effect, RFE/RL reported. Le Détroit's reading is that the warnings failed for three reasons that recur in every such case: they were private (a secret letter), late (October, when the rates were already exploding) and hedged (a distinction that gave the bigger schemes a certificate of respectability). Against an endorsement by the state, a confidential memo has no chance.

Part 8The run: Sude, Gjallica and the cash that left the banks

The first domino fell on 19 November 1996, when Sude defaulted. By mid-January 1997 riot police were using batons on protesting depositors, and its owner, a former shoe-factory worker, told them through a megaphone that the scheme had failed, RFE/RL reported. Sude and Gjallica declared bankruptcy, the other schemes stopped paying, and on 26 January the government froze the bank accounts of Xhafferi and Populli, holding about $250 million, Jarvis writes; Albania's envoy to the UN put the sum at about $300 million. VEFA, Kamberi, Silva and Cenaj cut their rates to 5% a month to reassure depositors. It did not work. In Vlorë, Gjallica's president announced on television on about 4 February that it had no money, RFE/RL reported.

The banks had seen the pyramid in their own books long before. By September 1996, VEFA's deposits alone made up more than a third of all demand deposits in the Albanian banking system, Jarvis notes. Bank of Albania data in his paper show private-sector demand deposits in lek quadrupling from 7.7 billion lek in September 1996 to 31.1 billion in December, as cash flowed in at the peak (the table does not identify the depositors). A single non-bank customer holding a systemic share of deposits is the kind of red flag a supervisor can see without any new law.

Then the money moved the other way. Cash outside the banks rose from 43.4 billion lek in November 1996 to 77.6 billion in July 1997, and the ratio of cash to bank deposits from 42.9% to 81.1% (Figure 6). Deposits fell from 106.7 billion lek in December 1996 to 95.7 billion in July: Albanians who had lost their savings in the pyramids trusted nothing they could not hold. The money that survived came back slowly. Households' lek time deposits at banks more than doubled during 1997, from 27.3 billion lek to 70.5 billion. Le Détroit's reading: the pyramids did not ruin the banks, which were too small to be ruined. They ruined trust in every promise of a return, and rebuilding it took the second half of 1997.

Part 9From bankruptcy to rebellion: Vlorë, the arms depots and Operation Alba

The pyramids' map became the rebellion's map. The Xhaferri foundation had worked mainly in the villages of Lushnjë, Fier and Berat, Shqiptarja reports, and it was in Lushnjë, on 24 January, that rioters set fire to the town hall, RFE/RL reported. In the same days, protesters in Lushnjë held the foreign minister, Tritan Shehu, in the town's stadium, the Albanian news site Balkanweb recounts; RFE/RL placed the episode in Vlorë. Gjallica was based in Vlorë. After it collapsed, more than 20,000 people reportedly took to the city's streets on 6 February and occupied its main square (Figure 7).

The state then came apart in a matter of weeks. Prime Minister Meksi resigned on 1 March and a state of emergency was declared on 2 March. In Vlorë, by then the centre of the revolt, protesters stormed police stations and military warehouses; Sarandë declared an autonomous government; in Gjirokastër army tanks met little resistance, and two pilots flew a MiG to Italy saying they had been ordered to fire on civilians, The Christian Science Monitor reported. On 10 March rebels took Berat and the Kuçova air base and came within 50 miles of Tirana; thousands of the army's 54,000 soldiers were believed to have defected, the same paper reported. On 11 March, Bashkim Fino formed an interim coalition, and the unrest spread north; arms depots were looted across the country. Estimates of the weapons taken vary widely, from more than 500,000 rifles to about a million.

The international response was fast once it came. On 28 March the UN Security Council adopted Resolution 1101, authorising a temporary multinational protection force for Albania under Italian command, by 14 votes to none, with China abstaining. Operation Alba began deploying on 15 April with about 7,000 troops, escorting humanitarian convoys and, later, election observers, and completed its withdrawal on 11 August. Elections on 29 June and 6 July gave the Socialist Party 101 of 155 seats; Berisha resigned on 23 July and parliament elected Rexhep Meidani president the next day. The risk that was most under-weighted beforehand, in Le Détroit's reading, was not financial at all. It was the state's capacity to absorb the anger of half its population on the same day.

Part 10The clean-up: what Albania and the IMF learned, and what they under-weighted

Albania eventually did most of the right things, in the wrong order. Its best decision was a refusal: the government made clear it would not compensate depositors from the budget, which Jarvis counts, with the freezing of Xhafferi's and Populli's bank accounts, among the two best things it did; freezing the accounts of the other schemes too, he adds, would have been better. Parliament banned pyramid schemes in February 1997 without defining them. A law of 9 May 1997 (No. 8215) introduced financial controls on non-bank legal persons, and a law of 30 July (No. 8227) let the government appoint administrators. Foreign administrators took up their duties only in November 1997 and gained full control in March 1998. For VEFA, Albanian administrators appointed in August 1997 were annulled by a court in September and reappointed in October, before an international firm was named on 24 November, the Strasbourg court's decision records. Seventeen companies went into administration.

Recoveries were uneven. Populli's depositors got back 60% of their money and Xhafferi's about half, from the deposits frozen in January 1997; for the schemes as a whole, assets would cover less than a tenth of nominal liabilities, Jarvis estimated. Even VEFA's debts are disputed: 85,000 creditors and 67.9 billion lek in Jarvis's table, 57,923 creditors and 29.2 billion lek in a 2001 expert report cited by the court, 73,509 creditors in a 2010 letter. Its owner, Vehbi Alimuçaj, was arrested in April 1998 and sentenced to 20 years on appeal in 2001; in 2012 the European Court of Human Rights found that sentence harsher than the law in force at the time of the offence, Top Channel reported. In 2016 a Tirana court gave him three years and eight months for hiding $328,000 in a Hungarian bank account, AP reported.

The banking system was rebuilt around the gap the schemes had exploited. New laws on the Bank of Albania (December 1997) and on banks (July 1998) were followed in March 2002 by a deposit insurance law creating a Deposit Insurance Agency; by then the central bank was licensing non-bank lenders, savings and credit associations and exchange offices, its 2002 supervision report shows. The state Savings Bank was sold to Raiffeisen in 2004. Jarvis's lessons still read well: build a working formal financial system; regulate informal markets and assign clear responsibility for them; investigate suspected schemes fast, with outside experts if needed; close them at once; freeze their assets; hand them to independent administrators; void recent transfers to related parties; rule out compensation from the start; and have the IMF and World Bank watch for schemes and press governments to act.

Three risks were under-weighted, in Le Détroit's reading, and they are the ones that matter most for fragile states. First, political economy: the lessons treat governance as a background condition, when endorsement by the state was the schemes' most valuable asset. Second, time: between the first default and the administrators' arrival a year passed, enough for assets to vanish. Third, the limits of the IMF itself. Jarvis conceded that the Fund cannot supervise individual firms and that public warnings may be ignored during a mania. That is true, and it is why the warning has to come from the domestic authorities, early and without qualification.

Part 11The pyramid goes online: Ponzi risk in fragile and transition economies today

Albania was not alone, and its conditions travel. In Romania, Caritas drew at least 2 million depositors, about a fifth of households, with promises of eightfold returns in three months before collapsing in May 1994, the anthropologist Katherine Verdery found. In Russia, MMM took money from millions before its bank lost its licence in 1994. In Serbia in 1992 and 1993, Jugoskandik and Dafiment paid 200% a month on dinar accounts in a sanctions economy with inflation around 25,000% a year, until they could not; two trade ministers were arrested. In Bitola, in what is now North Macedonia, the TAT savings house collapsed in 1997 with about 30,000 depositors and $60 million, and a deputy central bank governor was detained. Pyramids appear in rich countries too, Jarvis noted, but stronger enforcement and more sceptical investors keep them small (Figure 8).

The shopfront has since moved online; the arithmetic has not. MMM Nigeria promised 30% a month to a membership its founder put at 3 million, then froze accounts in December 2016, CNN reported. OneCoin, launched in Sofia in 2014 and sold through multi-level marketing as the "Bitcoin killer", took more than $4 billion from at least 3.5 million people, according to the US Justice Department. In Turkey, the Thodex exchange went offline in April 2021 with more than 400,000 investors; its founder was sentenced to 11,196 years and was found dead in his cell in November 2025, Turkish officials said. In Kazakhstan, the financial monitoring agency dismantled 24 pyramids in the first months of 2025, some taking crypto deposits, and blocked 5,600 fraudulent websites, ForkLog reported. Chainalysis estimates that crypto scams received at least $14 billion in 2025, with high-yield investment programmes among the dominant types.

Regulators know the pattern. A 2023 report by the FATF, Interpol and the Egmont Group of financial intelligence units describes cyber-enabled fraud as transnational organised crime whose proceeds move rapidly across borders, and calls for coordination between public and private sectors. In the EU, the MiCA regulation applies to many crypto-assets since December 2024, but the European supervisory authorities warned in October 2025 that protection varies by asset and service and may be limited, and that "finfluencers" on social media drive much of the interest. In Albania itself, the informal economy is still put at almost a third of GDP, and cash in it rose about 50% between 2018 and 2022, to €3.15 billion, the Global Initiative Against Transnational Organized Crime reported in 2024; the country left the FATF's grey list in October 2023.

Le Détroit's reading of the risk today: the economies most exposed are those that combine Albania's ingredients. Large remittance inflows and cash use; banks that pay little or lend little; non-bank and crypto products that no authority fully supervises; and politicians or celebrities willing to lend their names. Many transition economies of the Western Balkans and Central Asia, and most post-conflict states, tick several of those boxes. The schemes will not be called foundations this time. They will be called platforms.

Part 12Red zones: where the conditions rhyme today

No one can forecast a pyramid scheme. But the conditions that let a fraud grow into a national crisis can be measured, and Albania supplies the list: savings with no trusted bank to go to, remittances arriving as cash, prices that eat deposits, a supervisory gap the state did not close, and returns no honest business could pay. Le Détroit has turned that record into six echo conditions and screened 13 economies against them, using only values we could open and date (Figure 10).

Method. A screen of conditions, not a forecast; no country, firm or person is accused of anything. Each cell holds the latest public value, with its year and source in the tooltip; thresholds start from Albania's 1996 benchmark, when remittances were some 12% of GDP and inflation 12.7%. Ranks count two points per red cell and one per amber; a case with no published count is amber. A cell we could not source is marked "not assessed" and counts for nothing, so Kosovo, Lebanon and Vietnam may score lower than a complete record would show.

Egypt, Nigeria and Lebanon echo the most conditions, followed by Bosnia and Herzegovina, Pakistan, Turkey and Vietnam; Albania itself comes just behind, level with Uzbekistan. Three clusters stand out. The first is the cash economies, where most adults still have no account and money arrives from abroad. In Egypt, Pakistan and Lebanon, between 57% and 77% of adults had no account in 2024, against 54% in Albania itself; remittances were worth 11% of GDP in Egypt and almost 10% in Pakistan in 2025, and a third of GDP in Lebanon in 2023. That is Albania's 1996 profile almost line for line: a country with savings and no front door. Egypt adds inflation of 20% in fiscal 2024/25 and HoggPool, a crypto-mining app that took about $620,000 from thousands of investors before 29 arrests in March 2023.

The second cluster swaps the cash box for a phone. Nigeria and Turkey have more banked adults, but savings there lose value fast, with inflation of 23% and 35% in 2025, and crypto use is among the world's highest: Nigeria ranks third in Chainalysis's 2026 index, Turkey 20th. Each has had a recent collapse that echoes the Albanian pattern. CBEX, an app that promised to double deposits in 30 days, froze withdrawals in April 2025 and locked out thousands of Nigerians. In Turkey, the capital-markets regulator ordered 131 investment funds liquidated in September 2026 after two fund firms struggled to meet withdrawals, and prosecutors are investigating alleged Ponzi-type methods; no court has ruled. The funds, held by 455,758 investors, were worth about $18 billion, roughly 1% of GDP by our arithmetic.

The third cluster is closest to home. In Kosovo, Bosnia and Herzegovina, Kyrgyzstan and Uzbekistan, remittances are worth 11 to 18% of GDP, close to or above Albania's 12% in 1996, and the schemes recur in small, familiar shapes: an app paying for clicks in Bosnia in 2023, recruitment pyramids in Kyrgyzstan and Uzbekistan in 2024. In Uzbekistan a scheme called Ml Share recruits through "charitable" events, Kun.uz reported in February 2026: the foundations' disguise, thirty years on. And in Albania, police arrested six people over XueX, a crypto platform that asked each member to recruit at least five others, Shqiptarja reported, in September 2025, around Lushnjë and Divjakë in Fier county, where the Xhaferri foundation had collected its cash in 1996.

What differs from 1996 matters as much as what rhymes. No recent case we found comes near Albania's scale: the largest, in Turkey, is about 1% of GDP against 64%. Albania now has a deposit insurance agency, created in 2002. Supervisors elsewhere act, if late: Kazakhstan's monitoring agency closed 24 pyramids in a few months of 2025, Nigeria's new securities law makes Ponzi schemes a crime, and Vietnam's police seized assets worth over $197 million from the "Mr Pips" network. Digital rails cut both ways. In Kenya, 90% of adults now have an account, counting mobile money, yet CBEX reached Kenyans through Telegram and Quant Vest through a mobile app, which the market regulator blacklisted in September 2026. The rails also speed up the cycle: CBEX ran for about nine months. Albania's warning window was two years. Today's may be a season.

Indicators and warnings: what an analyst could have seen

A screen says where to look; the Albanian sequence says what to look for once there. Read as an early-warning file, the record holds ten signals before Sude's default, the first of them 23 months ahead (Figure 11). Four were in plain sight: the rates were printed, advertised and rising. The hardest evidence sat in bank and supervisory data, where VEFA's share of demand deposits and a doubling of private lek deposits in a single month gave about two months of unambiguous warning. The official warnings started early, in private; the public one came last, hedged. In a red zone today the order of watching is the same: the rate, then the bank data, then whether anyone in authority says no in public.

Part 13Le Détroit's reading

Albania's pyramid crisis is usually told as a story of naive people in a strange country. The record says otherwise. The schemes grew because a society with real savings had no functioning bank to put them in, and because the state, which knew, preferred to endorse rather than to close. Depositors acted on the information they had: rates printed in the newspaper beside the exchange rates, ministers on television, medals pinned at anniversary dinners. The fraud was in the schemes. The failure was in the institutions around them.

For banks and compliance teams, the lesson is that a pyramid shows up in their own data before it shows up in the news. A non-bank customer holding a third of the system's demand deposits; a fourfold jump in private-sector deposits in three months; fan-in from thousands of small payers; cash in four currencies: none of that needed a new law to see. Neither did the most important signal, which is not a high rate but a believable one that keeps rising. Eight per cent a month was never modest. It was 152% a year, promised by a company with supermarkets.

For supervisors, especially in transition and fragile economies, Albania argues for four rules. Warn in public, early and without distinctions between "good" and "bad" schemes. Give one authority the duty to act against anyone taking repayable funds from the public, whatever they call themselves. Freeze first and investigate second, because the assets leave faster than the administrators arrive. And treat political endorsement as a red flag in itself. The IMF learned that prevention is better than cure. The harder lesson is that prevention needs someone with the power, and the will, to say no while the scheme is still paying.