Money on a Spring

# Money on a Spring

How credit, asset revaluation and external inflows can grow faster than a country's production capacity — and why stopping that growth is dangerous

*All macroeconomic and monetary indicators are given as of October 2026 or for the latest published period. Calculations, definitions and sources are in the [statistical appendix](/en/article/money-on-spring-appendix-en).*

Uzbekistan today is one of the most talked-about economic stories in the CIS. The country seems to come straight out of "One Thousand and One Nights": trillions of soums, hundreds of tonnes of gold and growth rates many of its neighbours would envy. Only the numbers here are not fairy-tale — they are official. Over the nine months of 2026, nominal GDP grew by 17.4%, and the banking loan portfolio by 13.6% over the year. In September the M2 money supply posted a record monthly increase of 24.1 trillion soums. The gold reserve reached 446.2 tonnes by October 1, and the World Bank forecasts real economic growth of 7.9% in 2026 — one of the highest rates in the region.

The hype is not invented: the economy really is growing. The reason for this article is not to argue with that growth, but to note the coincidence of several processes that are usually considered separately. The Central Bank is already talking about a gradual reduction of rates in the coming years. The state is building a unified system of mass real estate valuation, one of whose stated purposes is to use the value of properties when they are pledged as collateral. Investment is rising fast, but a significant part of its financing is tied to foreign capital, and new enterprises do not always reach their design capacity.

These processes can reinforce one another. Cheaper credit expands borrowers' capacity; the new valuation infrastructure can change how collateral is handled; rising property values can support further lending. But financial expansion must be underpinned by real incomes, productivity, operating enterprises and foreign-currency earnings. If these grow more slowly, the outwardly healthy picture gradually accumulates vulnerability.

This is exactly what I want to test: where the money comes from, what it turns into, and how long the financial system can sustain acceleration if production, state institutions and people's skills fail to keep pace with the growing complexity of the economy. Where does rapid growth end and the accumulation of risk begin — for banks, enterprises, depositors and the state itself?

## The September record and what lies behind it

In September, Uzbekistan's money supply grew by 24.1 trillion soums in a month — by 5.5%. This concerns the M2 aggregate: it includes cash and bank deposits. At first glance it is simple: there is more money, so demand should accelerate, and prices should follow. But a single M2 figure does not yet say who created that money, whose hands it ended up in, or how quickly it will reach the market.

The September balance sheet has two lines without which the record looks quite different. Because of the gold revaluation, its contribution to the change in the money supply turned out to be negative — about 55 trillion soums. The Central Bank's capital shrank by roughly 66.5 trillion over the same month. In the M2 calculation, capital is counted among the items deducted from the monetary aggregate. When that value decreases, the deduction becomes smaller, and the line's contribution to the change in M2 becomes positive.

This is an accounting change, not a transfer of 66.5 trillion to someone's account and not new lending. In the end, the movement of gold and capital together added about 11 trillion soums to September's M2. Strip these two lines out of the calculation, and the monthly increase would have been about 13 trillion. Lending to the private sector added another 12.2 trillion, while government operations, by contrast, withdrew some of the money from circulation.

![Structure of the September increase in M2: credit, foreign assets, gold revaluation, Central Bank capital, government claims](/upload/upload-money-on-spring-fig1_m2_waterfall_en.svg)

Fig. 1. Decomposition of the increase in M2 in September 2026: credit to the private sector +12.2; foreign assets excluding gold +3.2; gold revaluation −55.3; Central Bank and bank capital +66.5; other lines and rounding +2.0; net government claims −4.5; total +24.1 trillion soums. The "gold + Central Bank capital" block contributes +11.2; without it the increase would be ≈ +12.9. The lines describe the balance sheet of the monetary aggregate, not actual expenditure. Data: Central Bank banking system review, as of 30.09.2026; calculations in the statistical appendix.

So the record figure cannot be read apart from the balance sheet. The money supply changes under the influence of bank credit, external operations, the budget and the revaluation of reserves. At the same time, money in an account does not necessarily turn into purchases right away: a significant part of it stays in deposits, while another part services construction, investment, trade and imports.

The main question is not how much money was added in a single month. What matters is what creates that increase, where the credit goes, and what real incomes must ultimately service the new liabilities.

**Finding 1. September's +24.1 trillion soums of M2 is not 24.1 trillion of new consumer demand. About 11.2 trillion came from the accounting block "gold — Central Bank capital", while credit to the private sector added 12.2 trillion. Even without the revaluation block, the money supply would have grown by roughly 12.9 trillion. The record is made up of revaluation and genuine credit expansion — the two must not be conflated.**

## Money, credit and trust in savings

Over the past twelve months, the broad money supply M2 grew by about 38.4% and reached 461.9 trillion soums by October 1, 2026. About 283 trillion — roughly 61% of M2 — was held in time and savings deposits in both currencies. In September, time and savings deposits in the national currency added 16.8 trillion soums.

A deposit is part of M2, but money in a time deposit is not as readily available for everyday spending as a balance on a card or current account. When a depositor moves money from a time deposit to a current account, M2 itself does not increase. Something else changes: the money becomes easier to use, and its velocity may rise.

A bank does not wait for a depositor to bring in money before passing it on to a borrower. When it issues a loan, it creates a deposit in the client's account. This is how endogenous money works: credit increases deposit money in the system. But savings still matter for funding costs, liquidity and the maturity of liabilities. Short-term and expensive resources sit poorly with long industrial loans.

So the gap between deposit and credit growth matters not as a ready-made diagnosis of the banking system, but as a question about the structure of its balance sheets: which loans are being issued, which are being repaid, in what currency and for what term banks raise money, how much liquidity they place in Central Bank instruments, and how many quality borrowers are ready to take on debt on acceptable terms.

According to the latest published Central Bank data, in August the average rate on household time deposits in soums was 18.4% per annum, against 20.6% in January. Across all time deposits in soums, the rate fell from 18.9% to 17.2% over the same period. The Central Bank's policy rate, by contrast, has remained at 14% since March 2025. At its meeting on November 4, the regulator left it unchanged, although a cut to 13.5% was discussed.

Deputy Central Bank Chairman Nodirbek Achilov said that lending and deposit rates would gradually move toward more balanced levels over the next three to four years. If inflation steadily approaches the 5% target, the policy rate, he said, could move into single digits in 2027–2029. For the regulator, this is part of the shift to cheaper money. For the depositor, it is a question of whether the deposit will retain its purchasing power.

Here a gap emerges that can be gauged from the Central Bank's own survey. In September 2026, official annual inflation was 6.4%. Households estimated the price growth that had already occurred at 9.2%, and their expectations for the next twelve months stood at 8.7%. In Tashkent, perceived price growth was estimated at 11.9%, and expected inflation at 11.6%. The CPI measures the average consumer basket, while the survey captures personal assessments and expectations. The gap shows that for residents of the capital, the official 6.4% diverges noticeably from the rise in expenses they actually feel.

For a depositor, what matters is not only the published inflation of the past year but also the expected rise in prices over the term for which the money is placed. The latest available average rate on household time deposits is 18.4% for August; the inflation expectations survey relates to September. Comparing these data using the Fisher formula, real returns at official inflation of 6.4% come to about 11.3%. At expected nationwide inflation of 8.7%, they fall to roughly 8.9%, and at Tashkent residents' expectations of 11.6%, to 6.1%. This is a scenario comparison of the average August rate with September inflation expectations, not the return on a specific deposit over a single period.

| Nominal deposit rate | With 8.7% expectations | With Tashkent's 11.6% expectations |
| -------------------- | ---------------------- | ---------------------------------- |
| 18.4%                | 8.9%                   | 6.1%                               |
| 14%                  | 4.9%                   | 2.2%                               |
| 12%                  | 3.0%                   | 0.4%                               |
| 10%                  | 1.2%                   | −1.4%                              |

The 14%, 12% and 10% rates are scenarios, not a forecast of bank offers. They show how quickly the cushion of real returns disappears if bank rates fall while inflation expectations stay high. A depositor compares the deposit with their own spending and with alternatives — foreign currency, gold, housing, or buying goods now.

![Nominal rate on time deposits of 18.4% against four measures of price growth; real return under the Fisher formula](/upload/upload-money-on-spring-fig2_rate_vs_inflation_en.svg)

Fig. 2. The average rate on time deposits in soums — 18.4% (August 2026) against four measures of price growth for September: official CPI of 6.4%, perceived price growth of 9.2%, expected inflation of 8.7% nationwide and 11.6% in Tashkent. Real return under the Fisher formula — ≈ 11.3%, 8.9% and 6.1%. A scenario comparison of the August average rate with September expectations, not a forecast. The Central Bank's policy rate has been 14% since March 2025. Source: Central Bank, inflation expectations survey; data in the statistical appendix.

That is precisely why distrust of the inflation measurement matters for monetary policy. If people believe prices will rise faster than the official index shows, a cut in the deposit rate can weaken the incentive to keep savings in soums. Yet it cannot be said that a mass exodus from the banks has already begun. In 2025 banks sold households 661 kilograms of gold bars — noticeably more than a year earlier — but their value remained small next to the deposit base. And in the nine months of 2026 households sold banks $18.9 billion of foreign currency, and the soum strengthened against the dollar by roughly 1.5% since the start of the year. This is the gross flow of currency sales, not the net change in households' foreign-currency savings: such a conclusion would require data on offsetting purchases, movements in deposits and acquisitions of other assets. For now these data show a substantial supply of foreign currency from households, but they do not give a complete balance of household savings.

But they do not answer the question about the future. If deposit rates fall faster than inflation expectations, will depositors' behaviour change? Some money may move into foreign currency, gold or housing; some into consumption. The last option matters most: savings do not simply shift between financial assets — they can reach the consumer market faster.

There are other large balances that should not be confused with household deposits. Treasury accounts and the accounts of budget organisations held about 304 trillion soums. That is roughly two-thirds of M2 by size, but the government balances themselves are not part of M2. While the funds sit in accounts, they do not support private demand; when the budget spends them, part returns to circulation. Yet the entire balance cannot be treated as a sum the state could direct into purchases at a single stroke.

The Central Bank's sterilisation liabilities to banks reached roughly 88 trillion soums by September. The regulator uses such instruments to lock up free liquidity. Over the past twelve months these liabilities grew by 70.5 trillion. This shows that banks are channelling a significant volume of liquidity into Central Bank instruments. At the same time, lending is distributed unevenly: as of August 1, 2026, lending to industry fell 13% year on year, while lending to construction rose 32%. This combination requires looking not only at the price of money but also at the quality of projects, the maturity of funding and the sectoral distribution of risk.

In January–September nominal GDP grew 17.4%, real GDP by 8.3%, and the GDP deflator was 8.4%; annual CPI in September was 6.4%. The deflator reflects the prices of all value added produced domestically, CPI reflects the consumer basket. They cannot be treated as interchangeable measures, just as M2 over twelve months cannot be directly compared with nominal GDP over nine.

So the money supply is growing fast, but its effect on demand is determined by the credit flow, the maturity of funding and depositors' decisions. The Central Bank faces a difficult task: to make credit gradually cheaper without devaluing the incentive to keep savings in soums.

**Conclusion 2. Savings still retain a positive expected real return: at a rate of 18.4% it is about 6–9%, depending on inflation expectations. But if the deposit rate falls to 12%, the cushion for Tashkent almost vanishes, and at 10% it turns negative. At the same time, sterilisation liabilities grew by 70.5 trillion soums, and bank credit is distributed unevenly. The question is no longer only how cheap money is, but whether it will stay in soums and reach productive projects.**

## Investment is growing. Can production keep up?

In January–September 2026, 584.2 trillion soums were invested in fixed capital. The official index of the physical volume of investment rose 20.1% compared with the same period a year earlier. The 584.2 trillion is the value of investment in current prices, while the 20.1% is growth in comparable prices. These are different measurements: the nominal sum invested and the real rate of its growth.

The largest share went to manufacturing — 160.8 trillion soums. Housing construction accounted for 87.1 trillion, agriculture for 52.1, construction for 51.1, energy for 39.0, and transport and storage for 27.7 trillion.

Under the official classification, 371.7 trillion soums, or 63.6% of investment, came from foreign investment and non-guaranteed foreign loans. A further 36.3 trillion, or 6.2%, came from foreign loans under state guarantee. Together that is about 69.8% of fixed capital investment. This is not the share of foreign direct investment, nor the share of external debt: the categories combine different instruments. But they show how dependent the investment cycle is on foreign capital and foreign currency.

Domestic commercial banks still play a comparatively small direct role in investment financing. According to the State Statistics Committee, commercial bank loans and other borrowed funds accounted for about 1.3% of the sources of fixed capital investment. This does not mean banks are not financing the economy: their loans also go to working capital, trade, construction and housing. But the great investment push does not in fact rest mainly on long-term lending by local banks.

Growth in the bank portfolio is also unevenly distributed. According to the Central Bank, as of August 1, 2026, the total loan portfolio grew by about 12% over the year. Loans to construction rose 32%, to trade and catering 25%, to individuals 19%, while lending to industry fell 13%. By a later overall estimate the portfolio reached 713.3 trillion soums by September 30, up 13.6% over the year. The contrast shows where credit activity is shifting: construction and trade are growing fastest, while industry is losing bank financing.

The main problem of the investment cycle is time. Money starts being spent long before any additional output appears. Land must be bought, equipment ordered, shipped, the site prepared, buildings put up, electricity and water connected, lines installed, workers trained and component supplies arranged. At every stage demand already arises for materials, transport, specialists and foreign currency. The product that is supposed to repay all these costs appears much later.

When many projects are launched at once, they compete for the same resources. Construction sites need cement, metal, cable, fuel and transport. Production facilities need electricity, water, specialists, raw materials, spare parts and services. Imported equipment and components are paid for in foreign currency. An investment boom can therefore raise imports and costs before new plants start turning out products.

In April 2026 the authorities reported that 210 of the 688 enterprises launched over the previous three years had not reached full capacity. The possible additional output from loading them was estimated at 33 trillion soums — about 1.8% of nominal GDP for 2025. The authorities cited the reasons: shortages of working capital, raw materials, equipment, connections to utilities and engineering infrastructure; in some cases project delays and weak demand got in the way. This is an estimate of potential output not produced, not a proven sum of losses or evidence that all the investment was wasted.

For a different group of enterprises, the possible additional output and exports were estimated separately in May — at up to 65.6 trillion soums. This is a different group and a different estimate. It cannot be added to the 33 trillion. But both figures point to the same question: how many of the launched projects are actually operating, and what is stopping the rest from turning into sustainable production?

NGMK shows why the financial result cannot be taken as a measure of physical output. In the first half of 2026, the combine mined 1.51 million troy ounces of gold — 1.8% less than a year earlier. Revenue nevertheless rose 49.9% to $7.07 billion, EBITDA by 59.8% to $4.89 billion, and net profit by 57.2% to $2.38 billion. The main factor is the high world price of the metal: the combine received considerably more money even though it physically mined slightly less.

Output is constrained by the gold content of the ore, stripping operations and plant throughput. Total cost including sustaining expenses rose 41.5% to $1,647 per ounce: NGMK attributed this to royalties, materials and expanded stripping at Muruntau.

Profit and cash flow diverged: EBITDA rose 59.8%, while net operating cash flow fell 3.4%, from $2.60 billion to $2.51 billion. Accounting profit is therefore not equal to the sum available for new projects.

The picture with local suppliers is also mixed. In the first half of the year, output under NGMK's localisation programme rose 6.8% in value terms, and purchases under inter-industry cooperation by 15.1%. These are different programmes, and neither covers all of the combine's procurement. They do not show what share of the value of output is actually created inside the country and what share falls to imported parts and materials.

Imports mix equipment, raw materials, components and consumer goods. In January–July 2026, machinery and transport equipment accounted for $9.7 billion (+20%), but this category also includes passenger cars. Without a commodity breakdown, it is impossible to determine what part of imports directly expands production capacity.

Importing raw materials and components is natural for a growing economy. The question is the share of local value added: if a plant constantly buys key parts and technologies abroad, gross output grows faster than the domestic production base.

There is also consumer imports. In January–July, food and live animals were imported to the value of 3.3 billion dollars — 41.9% more in dollar terms. In January–September, retail trade grew 19.7% in real terms, while agriculture, forestry and fishing grew 4.6%. Within the agricultural sector, dynamics varied: meat production rose by roughly 2.1%, milk by 1.7%, eggs by 7.5%.

Imports are measured in dollars and agricultural output in real terms; commodity coverage also differs. But the combination of fast retail growth and food imports with slower growth in agricultural output calls for a commodity-by-commodity check: what exactly can local production not supply in the required volume, quality or at the required price?

Domestic demand is supported by wages and remittances. In the first half of 2026, remittances reached $9.3 billion (+13%), and in the first quarter the Central Bank noted real wage growth of 9.5%. Over nine months, inbound remittances totalled $15.17 billion against $13.93 billion a year earlier, but the growth rate slowed from 23.7% to 8.9%. The positive balance reached $13.2 billion. These inflows support demand, although part of the money stays in accounts or goes into assets.

Why does domestic production not always displace imports? In some cases seasonality gets in the way; in others there is a shortage of raw materials, processing, packaging, storage or logistics. In other cases imports are cheaper, more consistent in quality or delivered faster to retail chains. For food, a commodity test is needed: for each group, compare physical imports, local output, exports and changes in stocks. Dollar totals alone are not enough.

For capital and intermediate goods there are several further obstacles. A local enterprise may receive orders only for the duration of construction of a large facility, and once the project is completed the orders disappear. The next facility may require different standards, parts, certificates and service. A foreign supplier often wins through scale and an established service network.

The second factor is productivity. According to an OECD estimate, labour productivity at companies with foreign participation in Uzbekistan is roughly 63.8% higher than at domestic firms. This does not prove that a foreign manager is inherently more efficient: such companies may be larger, more capital-intensive and operate in different industries. But the gap is a reminder that registering an enterprise and putting money in does not yet make it competitive.

The third factor is how access to the market is structured. If contracts, subsidised loans and infrastructure are allocated with regard to administrative proximity or personal connections, an inefficient supplier can remain shielded from competition for a long time. An entrepreneur then finds it more profitable to cultivate relations with those who allocate resources than to invest in technology, quality and productivity. This does not mean that every decision is made through connections or that the entire economy is organised this way. But such a hypothesis requires testing through data on the ultimate owners of companies, related parties in procurement, repeat tender winners, subsidies and contract performance results.

The Audit Chamber reported violations of legislation in public procurement worth 9.1 trillion soums in the first half of 2026. These are identified violations, not proven damage or the amount of inflated contracts. But together with the productivity gap and the temporary nature of some orders, they show why the selection of projects and suppliers must rest on verifiable efficiency.

**Conclusion 3. The investment boom is almost 70% financed by the categories of foreign investment and foreign loans, while commercial bank credit and other borrowed funds provide about 1.3% of the sources of investment. At the same time, 210 of 688 recently launched enterprises have not reached full capacity. Output and GDP are growing, but the conversion of investment into sustained output, local value added and competitive enterprises remains incomplete.**

## When an asset becomes collateral, and collateral becomes the source of a new loan

Now to real estate. Here it is important to distinguish three things: the market price of a property, its cadastral or mass appraisal, and the value a particular bank is prepared to recognise as collateral. They are related, but they do not automatically coincide.

In March 2025, Presidential Decree No. UP-43 introduced a system of mass appraisal of real estate. The document explicitly names its goals: to determine the economic value of more than eight million residential and non-residential properties and roughly seven million land plots; to ensure the consistency of valuations used for taxation, purchase and sale, pledging as collateral and other transactions; and to create a regularly updated database of the real estate market. In other words, using valuations for collateral is not an accidental side effect of the reform but one of its officially stated objectives.

According to preliminary results of the mass appraisal in Tashkent, more than 600 thousand apartments received a combined valuation of 426.7 trillion soums against the previous cadastral base of about 57.8 trillion. More than 213 thousand private houses were valued at approximately 548.4 trillion against the previous 39.1 trillion. In the first case the new figure is about 7.4 times the old one; in the second, about 14 times.

The ratios of 7.4 and 14 times show the difference between the new mass appraisal and the previous cadastral base, not an increase in market prices. The residential real estate price index in the third quarter of 2026 showed annual growth of about 4.8% nationwide and 5.1% for apartments in Tashkent. Market prices are rising moderately; the multiple gap arose from the revision of the old administrative valuation.

But here it is easy to draw a mistaken conclusion: that before the mass appraisal, banks determined collateral value from the old cadastral figure. That is not so. The cadastral record and the valuation of a specific property for a transaction or loan are different things. Independent appraisal organisations have long operated in Uzbekistan. The owner or the bank could commission an individual valuation: a specialist studied the documents and characteristics of the property, inspected the property where necessary, and the organisation then issued a report. If the valuation was carried out without an inspection, the national standard required the reasons and the resulting limitations to be stated.

This procedure existed long before UP-43. The Law "On Mortgage" provided that the value of the mortgaged item is determined by agreement between the mortgagor and the mortgagee or by valuation under the rules of legislation on appraisal activity. And paragraph 36 of Central Bank Regulation No. 3269, in force until July 24, 2025, allowed a bank to calculate the loan-to-value ratio on the basis of an appraisal organisation's report or the price stated in the purchase and sale contract. In other words, the bank was not obliged to rely on the old cadastral value: for a specific loan it could use an individual market valuation or the transaction price.

Comparing 426.7 trillion with 57.8 trillion and 548.4 trillion with 39.1 trillion therefore cannot be read as a 7.4-fold or 14-fold increase in the value of bank collateral. It shows how far the new mass valuation differs from the old cadastral base. A market valuation for banking transactions existed before as well.

The novelty of the reform lies elsewhere: the state is creating a single, regularly updated mass database that is meant to align property valuations for taxation, purchase and sale, pledge and other transactions. Such a database could make valuation more uniform and accessible to different market participants. But its effect on lending depends on exactly how banks apply it: as an additional reference point, as a source of input data, or as a replacement for some individual valuations. The mere existence of the database does not mean that the collateral value recognised by banks has automatically risen.

A mass valuation does not by itself turn into a bank's liquidation value. The bank still assesses the asset's liquidity, legal cleanliness, the likely discount and the borrower's ability to service the debt. For ordinary retail mortgages, the baseline macroprudential LTV limit is 80%, with limited exceptions; this is not a universal cap for any property-backed loan. So a sevenfold ratio of the new mass valuation to the old cadastral base does not mean the loan can be increased sevenfold.

If banks begin to use the new base widely when valuing existing properties, it could simplify procedures, unify input data and in some cases affect the recognised value of collateral. Some owners would then have more room to refinance or take out additional credit. But that effect needs to be tested against actual bank valuations and loan disbursements, not derived from the difference with the old cadastral base. Against a backdrop of gradually falling rates, such a channel could strengthen demand for property.

This is where a loop forms that is familiar from other countries. First, the value of the property is revised or rises. A higher valuation lowers the debt-to-collateral ratio. The borrower gains the ability to borrow more. The additional credit supports demand for property. Deals are struck at higher prices, which become a new reference point for appraisers and lenders. The next round begins with a larger collateral value.

This is precisely how a revaluation can stop being a simple statistical procedure and become part of the financial cycle. A higher value for an existing asset does not by itself create a deposit or increase M2. But a loan issued against that asset does create a new deposit and additional purchasing power. What matters, therefore, is not only the fact of the revaluation but also how actively additional loans will be issued against the new valuation.

The official goal of the reform is to bring property valuation closer to market levels, create a single database and use a consistent value for different transactions. The state seeks to accelerate economic growth and demonstrate rising prosperity. The risk arises not from a hidden intention but from the combination of instruments: the revaluation widens the potential collateral base, while lower rates make borrowing cheaper. If productivity and incomes lag, the asset market may receive a credit impulse before the economy creates new sources of debt repayment.

Hyman Minsky's financial instability hypothesis explains this mechanism well. At an early stage, a borrower services debt from the income of its own operations. Then the ability to refinance the debt becomes increasingly important. In the extreme case, debt service depends on a new loan or on a further rise in the asset price. Minsky called such positions hedge, speculative and Ponzi finance. These are not labels for individual industries but ways of describing where a borrower expects to get the money to pay its debts.

The problem arises when growth in collateral value begins to replace growth in cash flow. An apartment may rise in price, but its owner has not necessarily started earning more. An enterprise may receive a higher property valuation, but its products have not necessarily become more competitive. If a loan is serviced only out of rent, profit or wages, the revaluation helps only within the limits of future income. If instead the loan is sustained mainly by the expectation of a further rise in collateral value, the system becomes sensitive to a market reversal.

The comparison with the US is useful if the differences are kept in mind. According to the Fed's financial accounts, from the first quarter of 2020 to the second quarter of 2026 the increase in the net worth of households and nonprofit organisations on one measure was about 86.7 trillion dollars; roughly 71.9 trillion came from the revaluation of assets that already existed, rather than from new savings and transactions. A revaluation can create a sense of rapid enrichment without meaning that new goods, capacity or cash flows worth the same amount have appeared. But the US issues the world's reserve currency and has a deeper financial market. For Uzbekistan the same process is complicated by external currency dependence.

**Conclusion 4. UP-43 explicitly included the pledging of property among the goals of the mass valuation. The gap between the new and old bases — 7.4 times for apartments and about 14 times for houses — does not imply the same growth in market prices or an automatically available loan. But the state is creating infrastructure that could expand the recognised collateral value of existing assets. Against a backdrop of falling rates, this is a potential channel for a credit-collateral loop: its scale will be determined by how banks begin to apply the new valuation.**

## The economy is becoming more complex faster than institutions and skills

Financial and production systems become more complex as an economy grows. A large project is no longer reducible to the relationship between an investor, a bank, a builder and a buyer. It involves several lenders, appraisers, contractors, subcontractors, importers, customs, certification bodies, energy and transport companies, parts suppliers, insurers and government agencies. Each participant depends on the timing and decisions of the others.

If one link is delayed, the idle time spreads further. Equipment may arrive but not get a grid connection. A plant may be built but be left without working capital or the necessary raw materials. A supplier may win an order on one construction site but fail the requirements of the next project. The more such links there are, the more expensive coordination errors become and the higher the demands on management quality.

The state itself acknowledges the problem of excessive procedures. In May 2026, during discussion of the "Eliminating Bureaucracy — 2030" programme, the authorities reported that government bodies perform 5,650 functions, more than 42 thousand mandatory requirements regulate business, and citizens have access to 1,041 types of public services. The statement separately noted paperwork, duplicative requirements and procedures. Some functions are necessary for security, control and the protection of rights, but excessive and overlapping requirements raise the cost of approvals and slow projects down.

A growing economy needs not only digital services but also professional coordination. Decisions must be taken by competent specialists, agencies must exchange data, responsibility must be clear, and rules must not change unpredictably. When formal procedures work poorly, people start to bypass them through acquaintances, personal connections and informal arrangements. At the everyday level this sometimes helps resolve an individual issue. At the level of a complex economy, such a system can entrench privileged access to contracts and resources, hamper competition and keep inefficient suppliers in place.

This is the economic side of cronyism. The problem is not that people support relatives or trust acquaintances. The problem arises when a personal connection substitutes for professional selection, verification of results and equal access to the market. Enterprises then compete not so much on productivity and quality as on proximity to those who allocate credit, land, infrastructure and government orders. Establishing the scale of such allocation requires data on ultimate owners, connected bidders, repeat winners and project outcomes. The economic mechanism is clear: if access to resources depends on relationships, cultivating connections may prove more profitable than investing in productivity.

The second part is human capital. A modern enterprise needs more than equipment. It needs engineers, technologists, setup technicians, power specialists, and experts in quality, logistics and management. A worker must be able to read technical documentation, work with measurements, identify errors and master new processes. A manager must calculate costs, assess risks and check whether an investment delivers results.

One available benchmark is the international PISA study. In the latest published cycle, PISA-2022, only 19% of fifteen-year-old schoolchildren in Uzbekistan reached the baseline level in mathematics, 14% in reading and 19% in science. The OECD averages were 69%, 74% and 76% respectively. These are the results of schoolchildren who took the test in 2022, not an assessment of all current workers or entrepreneurs. But they show the scale of the task: the country will have to expand basic skills substantially, without which it is difficult to master complex technologies and work with data on a mass scale.

There is also a broader question — the culture of handling knowledge. In the digital environment one can find videos in which the authors explain physical laws, medicine or the workings of the world through their own interpretation of religious texts and present this as scientific proof. This is not about Islam, nor about religiosity as such. Theology, personal faith and the natural sciences answer different questions and use different methods. The problem begins when the authority of the speaker replaces experiment, measurement and the possibility of verifying a claim.

The same applies to financial gurus, political bloggers and pseudo-experts. A person accustomed to accepting a persuasive explanation without checking it is vulnerable not only to pseudoscience but also to promises of easy returns, dubious investments and simplified economic theories. A 2023 SWP study by Andrea Schmitz describes the influential online environment of religious preachers in Uzbekistan; a 2025 qualitative study of the blogosphere examines individual channels and their subject matter. These works do not measure the prevalence of videos in which religious interpretation is passed off as proof of physical laws, nor do they link them to productivity. This example therefore matters not as an explanation of economic problems but as part of a broader question: how well society can distinguish authority from evidence.

Financial expansion can move fast. A new valuation base for real estate can be created by administrative decision; credit conditions can be changed; an investment programme can be expanded. The qualifications of workers, engineering culture and trust in formal rules take far longer to form. If the economy becomes more complex faster than these capabilities develop, bottlenecks arise: the money has been invested, the equipment purchased, but the launch is delayed; the project is built but does not reach its designed performance; the loan is issued but the company's cash flow does not cover the debt.

In that case the problem is not reducible to a shortage of money. The system cannot turn it into results fast enough. For an economy that wants to move from construction and importing equipment to producing on its own, upgrading technology and exporting, this becomes a constraint in its own right.

**Conclusion 5. The growing complexity of the economy outpaces the ability of institutions and of some enterprises to coordinate projects: 5,650 state functions, more than 42 thousand mandatory requirements for business and identified procurement violations worth 9.1 trillion soums show the scale of the administrative burden and of selection risks. These figures do not prove that all resources are allocated through connections, but they show why transparent project selection and professional accountability become a matter of productivity, not just of governance.**

## Gold and the currency ceiling

Gold plays two different roles in this story. Non-monetary gold is counted as commodity exports: when a foreign company buys it, the country receives foreign currency. Monetary gold is a reserve asset of the Central Bank. When the Central Bank buys the metal from a local producer for soums and keeps it in reserves, soum liquidity arises in the banking system, but no foreign buyer transfers currency for that batch. Buying gold can support the domestic money supply, but by itself it does not increase the available flow of dollars.

For January–September 2026, imports amounted to 39.6 billion dollars (+15.7%), exports to 25.8 billion (−5.5%), and the negative balance to about 13.7 billion. At the same time, exports of goods excluding gold rose to 13.3 billion dollars (+31.6%), and exports of services to roughly 9.7 billion (+31.1%). Exports of non-monetary gold fell from 9.8 billion dollars in the nine months of 2025 to 2.8 billion in the same period of 2026. Almost the entire deterioration in the foreign trade balance is therefore tied to the gold line, not only to the acceleration of imports.

The world market price of gold rose by about 42% over the comparable period. The estimated physical equivalent of non-monetary exports was about 95 tonnes in the nine months of 2025 and about 19 tonnes in the nine months of 2026. This is export value divided by the average price, not actual customs tonnage. NGMK produced 72.99 tonnes in the nine months of 2025 and reported about 2.3 million ounces — roughly 71.5 tonnes — in the nine months of 2026. Almalyk MMC publishes its fulfilment of the refined gold plan but does not disclose a comparable tonnage. The country's total output for the nine months therefore cannot be reconstructed from public data.

The Central Bank's annual balance sheet for 2025 shows that the regulator bought 109 tonnes of gold from local producers and sold 100 tonnes abroad. The physical stock increased by about nine tonnes. The last known foreign shipment came in September 2025 and amounted to about 1.5 billion dollars. In 2026 the direction of operations changed: according to the World Gold Council and the Central Bank, the gold stock has increased by about 55 tonnes since the start of the year and reached 446.2 tonnes by October 1.

If the Central Bank did indeed not resume foreign shipments after September 2025, the remaining exports of non-monetary gold in 2026 probably come from direct sales by producers. But there is no official breakdown, so this remains an explanation rather than an established physical balance.

Buying gold from local producers creates soum liquidity. If the metal stays in reserves, it brings no currency from a foreign buyer. The Central Bank has said that the additional liquidity from monetary gold operations was sterilised through foreign exchange operations and other monetary policy instruments. The published data contain several quantities that cannot be added together: about 41 tonnes of net purchases in the first half at a conditional price of 4,500 dollars per ounce corresponds to roughly 5.9 billion dollars; the change in the Central Bank's external balance sheet position over twelve months was about minus 4.4 billion dollars; sterilisation liabilities over the same twelve-month period grew by about 5.8 billion dollars in converted terms. These are different periods and different balance sheet items. No precise sum of foreign exchange interventions can be derived from them.

![Gold flows and their monetary traces: producers, Central Bank, reserves, exports, domestic market, sterilisation](/upload/upload-money-on-spring-fig3_gold_flow_en.svg)

Fig. 3. Gold flows and monetary traces, 2025 — September 2026: the Central Bank bought 109 t from local producers in 2025 and added 55 t in the nine months of 2026 — this creates soum liquidity but brings no currency; the last foreign shipment was September 2025 (~$1.5 billion) and was not resumed in 2026; non-monetary exports in the nine months of 2026 — ≈ 19 t (estimated equivalent) for $2.8 billion; the domestic market (UzRTSB) — 8.84 t, about 90% bought by jewellery enterprises; Central Bank sterilisation liabilities — +70.5 trillion soums over 12 months. Gold stock — 446.2 t as of 01.10.2026. Sources: Central Bank, the State Statistics Committee, WGC, Interfax, Gazeta.uz, Kursiv; data in the statistical appendix.

For January–September, the domestic foreign exchange market received 9.5 billion dollars from exporters' sales and 18.9 billion from the population. Demand for currency amounted to 50.4 billion, and supply excluding Central Bank operations to 43.8 billion. The gap reached 6.6 billion dollars against 7.79 billion in the same period a year earlier. As of October 1, the foreign currency component of international reserves stood at 5.92 billion dollars. The gap between demand and supply is a flow over nine months, while the foreign currency component of reserves is a stock on a single date. These are different measures, but both show that domestic soum resources do not substitute for foreign currency for imports and external payments.

The soum is an independent national currency with a floating exchange rate. It cannot literally be called a derivative of the dollar. But the Central Bank can create soums, whereas it cannot create dollars in that way to pay for imports and service dollar loans. Foreign currency is brought in by exports of goods and services, remittances, investment and borrowing. If inflows fall short, the difference has to be covered from reserves, raised on the market or met by reducing demand for imports.

The risk is therefore especially high for companies that borrow in dollars but earn revenue in soums. When the soum weakens, their debt in the national currency rises, and imported components and equipment become more expensive. Exporters with foreign-currency revenue are partly protected by natural hedging. The IMF notes that a significant share of foreign-currency loans goes to corporate borrowers without natural or market protection against exchange-rate movements; according to the fund's estimate, about 20% of companies do not earn enough profit to cover interest expenses.

If collateral prices fall and the soum weakens at the same time, the banking system takes a double hit. The value of collateral declines while the soum burden of foreign-currency liabilities grows. Banks tighten lending and demand additional collateral; companies postpone investment, cut costs or try to refinance old debts. For a household with a soum mortgage, devaluation does not increase the nominal outstanding debt, but it can reduce real incomes and raise the cost of goods.

**Conclusion 6. The external limit is set not by the quantity of soums but by the availability of foreign currency. Exports of non-monetary gold fell from $9.8 billion to $2.8 billion over nine months, while imports reached $39.6 billion. The gap in the domestic foreign-exchange market was $6.6 billion; the foreign-currency portion of reserves was $5.92 billion. These are not comparable measures of "deficit" and "stock", but they show the strain on the external circuit. Soums can be created inside the country; dollars for imports and for servicing foreign-currency debt can only be earned, attracted or spent from reserves.**

## How financial growth turns into vulnerability

A credit boom on revalued assets, fed by external financing and weak project selection, is a familiar construct. Japan went through it in the late 1980s, Thailand on the eve of the 1997 crisis, Spain before the 2008 crisis. The mechanics differed in each country, but all three cases show how rising asset values can amplify credit and then make the downturn more painful.

| Element of the construct | Japan, 1985–1991                                                                                                                                   | Thailand, 1995–1997                                                                                                           | Spain, 2000–2007                                                                             | Uzbekistan, 2026                                                                                                                                                                                                                              |
| ------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Catalyst                 | Yen appreciation after the Plaza Accord, monetary easing and financial liberalisation; banks sought new lending outlets                            | Opening of the financial sector and capital inflows under the baht's dollar peg                                               | Low real rates after euro entry, expansion of mortgage and construction lending              | UP-43: mass property valuation with the explicit aim of unifying values for collateral; the Central Bank signalled a gradual reduction of rates as inflation slows                                                                            |
| Collateral channel       | Land secured corporate loans; rising land prices increased collateral capacity                                                                     | Real estate and developers' projects secured bank and financial-sector loans, including foreign-currency ones                 | Housing and land secured mortgages and loans to developers                                   | The new valuation base covers collateral operations; the ratio to the previous cadastral base is 7.4 times for apartments and about 14 times for houses, but it does not mean an automatic increase in bank valuations in the same proportion |
| External circuit         | The main mechanism was domestic bank lending against real estate; this is not analogous to Thailand's reliance on short-term foreign-currency debt | Short-term external inflows, a current-account deficit and corporate borrowing in foreign currency without sufficient hedging | External wholesale funding and interbank flows within the euro area                          | About 69.8% of investment is foreign investment and foreign loans of various types; some companies have unhedged foreign-currency liabilities                                                                                                 |
| Construction and assets  | Sharp rise in land and equity prices; then an equity-market reversal and a fall in land prices                                                     | Overheating of real estate and finance companies; after the reversal, defaults and bank problems                              | Construction employment rose by 47% versus 27% for the economy                               | Real construction growth of 12.3%; housing investment of 87.1 trillion soums; construction lending up 32% as of August 1                                                                                                                      |
| Credit and scale         | Bank lending against real estate grew rapidly; land values reinforced the collateral behind new loans                                              | Rapid credit growth, especially through finance companies and real estate                                                     | Private credit to GDP almost doubled in 2000–2007                                            | Credit/GDP of about 33.9%; the available series does not yet show a comparable doubling of leverage                                                                                                                                           |
| Institutional bottleneck | Concentration of bank risk in real estate and late recognition of the bubble                                                                       | Weak supervision of finance companies and delayed restructuring                                                               | Cajas: weak governance, concentration of loans in real estate and unstable wholesale funding | 5,650 functions of state bodies, more than 42 thousand requirements for business and procurement violations worth 9.1 trillion soums; these are measures of burden and risk, but not direct proof of cronyism                                 |

The table compares mechanisms; it does not declare the countries identical. In Japan the centre was domestic bank lending against land and equities. In Thailand real estate was combined with short-term external foreign-currency debt and a fixed exchange rate: once reserves were no longer sufficient to defend the baht, currency pressure quickly turned into a banking crisis. In Spain the boom was fed by cheap euro-area funding; a significant part of the risk was concentrated in loans to developers and households, especially at regional savings banks.

For Uzbekistan the sequence matters: valuing an asset raises the potential amount of collateral; credit allows the purchase of more assets; transactions lift price benchmarks; the next valuation locks in a higher base. If credit gets cheaper faster than incomes and productivity grow, prices begin to support themselves. This is Minsky's credit-collateral loop.

Uzbekistan is not yet repeating the Thai scenario. The country has a floating exchange rate, gold reserves stood at 446.2 tonnes as of October 1, and the population sold banks $18.9 billion of foreign currency over nine months. Deposits give banks domestic funding. But these are different buffers: gold still has to be monetised; foreign-currency sales by the population are a gross market flow, not the state's stock of dollars; soum deposits do not repay companies' foreign-currency obligations. A buffer buys time and flexibility, but not immunity from currency mismatch.

Fisher described debt deflation: mass asset sales lower prices but increase the real burden of the remaining liabilities. Minsky showed how prolonged stability gradually moves borrowers from servicing debt out of income to dependence on refinancing and on rising collateral values. Steve Keen developed models linking bank credit, debt dynamics and aggregate demand. Together with Robert Ayres and Russell Standish, he studied the role of energy in production. Krutakov, in his book "Oil and the World", offers a historical interpretation of the link between energy resources, credit and growth possibilities, drawing on a broader tradition rather than being the primary source of these ideas.

Energy sets the physical side of the cycle. In the nine months of 2026, Uzbekistan generated 69.6 billion kWh of electricity (+8.4%), consumption rose by 6.3% to 68.1 billion kWh, and imports fell by 27.2%. Generation is expanding, but assessing the buffer requires data on peak loads, grids and the capacity available to new enterprises. Financial expansion can accelerate demand for energy, but it does not replace it or remove infrastructure constraints.

Uzbekistan already has some elements of the described construct: revaluation of real estate for collateral operations, a policy of lowering rates, rapid growth in investment and construction, a high share of foreign sources and unhedged foreign-currency liabilities at some companies. But credit/GDP is about 33.9%, and the housing price index rose by roughly 5% year on year in the third quarter. This is not the picture of a bubble that has already happened. It is a reason to watch whether credit and collateral values are starting to grow faster than incomes.

What needs checking is new lending against existing real estate, LTV at origination and after revaluation, the ratio of housing prices to rents and incomes, corporate cash flow relative to debt service, the currency structure of liabilities, utilisation of new capacity and local value added. If these indicators deteriorate at the same time, revaluation stops being a statistical procedure and becomes a source of systemic vulnerability.

Breaking out of the cycle requires three restraints: mass valuation must not automatically turn into more credit without an independent appraisal of collateral; deposit rates must not lose real returns faster than inflation expectations fall; and projects and state support must be selected on actual cash flow, productivity and capacity commissioning, not just on funds disbursed.

**Japan, Thailand and Spain show how asset revaluation, credit and external funding can turn rapid growth into a debt correction. Uzbekistan has not yet travelled that path, but it is already building part of its infrastructure. The spring is compressed not by a single M2 figure but by the gap between the value of collateral and the incomes that must service the debt. The rupture will show up in bank balance sheets, in companies' foreign-currency liabilities and in depositors' ability to preserve their savings in soums.**

## Sources and methodological notes

### Money and prices

*   [Central Bank: monetary aggregates](https://cbu.uz/en/statistics/dks/4498375/), as of 30.09.2026.
*   [Analysis of the September increase in M2 and its components](https://bankers.uz/news/3573), 16.10.2026; data based on Central Bank statistics. M2 reached 461.9 trillion soums, a monthly increase of 24.1 trillion (+5.5%), and a year-on-year increase of 38.4%.
*   [State Statistics Committee: GDP for January–September 2026](https://stat.uz/en/press-center/news-of-committee/70889-gross-domestic-product-of-the-republic-of-uzbekistan-for-january-september-2026), 30.10.2026; real growth 8.3%, current GDP 1,755.5 trillion soums.
*   [GDP deflator for January–September 2026](https://www.uzdaily.uz/en/uzbekistans-gdp-grows-83-in-january-september-2026-30-10-2026/), 30.10.2026; nominal growth of about 17.4%, deflator 8.4%. This is not an alternative CPI.
*   [Central Bank: inflation indicators](https://cbu.uz/en/monetary-policy/annual-inflation/indicators/), September 2026: CPI 6.4%, core inflation 5.5%.
*   [Central Bank: weighted average deposit and loan rates](https://cbu.uz/en/statistics/rates/), August 2026: the average rate on time deposits of individuals in the national currency was 18.4%; in January it was 20.6%.

### Assets, collateral and currency risk

*   [Ministry of Economy and Finance: preliminary results of the mass valuation of real estate in Tashkent](https://gov.uz/en/imv/news/view/224350), 14.08.2026; more than 600,000 apartments were valued at 426.7 trillion soums against a previous cadastral value of 57.7 trillion; more than 213,000 private houses at 587 trillion against 39 trillion. This is a preliminary mass valuation, not an observed increase in transaction prices.
*   [Central Bank: loan-to-value (LTV) ratio](https://cbu.uz/ru/financial-stability/macroprudential-policy-tools/loan-to-value/), updated 10.03.2026; the general limit for mortgages to individuals is 80%, with limited exceptions and a separate regime for certain refinanced mortgages.
*   [State Statistics Committee/SIAT: residential real estate price index, year on year, Q3 2026](https://siat.stat.uz/data/4726/?lang=en), updated 28.10.2026; the nationwide index was 104.8, apartments in Tashkent 105.1.
*   [IMF: Republic of Uzbekistan — 2026 Article IV Consultation, Staff Report](https://www.elibrary.imf.org/view/journals/002/2026/152/article-A001-en.xml), 2026; deposit base +33.1% y/y in March, a declining loan-to-deposit ratio, credit growth in microcredit and mortgages, foreign-currency loans to companies without natural or market hedging, and about 20% of firms with insufficient profit to service interest.
*   [IMF: Informational Annex — exchange rate arrangements](https://www.elibrary.imf.org/view/journals/002/2026/152/article-A002-en.xml), 2026; Uzbekistan's de facto regime is classified as floating, with the exchange rate set on the foreign exchange market.
*   [IMF: Debt Sustainability Analysis](https://www.elibrary.imf.org/view/journals/002/2026/152/article-A003-en.xml), 2026; a significant share of public external debt is raised from official creditors on long maturities and on concessional/semi-concessional terms.
*   [Central Bank: Financial Stability Report for H1 2025](https://cbu.uz/upload/medialibrary/fea/zc1m2epsxmcmn8u19j8q8n4so1d7eqsz/Financial-Stability-Report-for-H1-2025.pdf), November 2025, sections on the real estate market and the macro stress test; housing price shocks of −30% and −50% and their impact on LTV and banks' capital adequacy are modelled.
*   [AIERA Frontiers: "What will happen to $84 trillion of American wealth"](https://aierafrontiers.com/ru/article/chto-budet-s-84-trln), 01.11.2026; a mechanism comparable in concept — the revaluation of existing assets within the structure of US household wealth growth — but in a different currency and institutional context.
*   [IMF COFER: the dollar's share of global official foreign exchange reserves in Q2 2026 was 56.7%](https://data.imf.org/en/news/imf%20data%20brief%20september%2030), 30.09.2026.
*   [The Fed: The International Role of the U.S. Dollar — 2025 Edition](https://www.federalreserve.gov/econres/notes/feds-notes/the-international-role-of-the-u-s-dollar-2025-edition-20250718.html), 18.07.2025; the dollar's role in international bank claims, liabilities and foreign-currency debt.

### Investment, output and capacity utilisation

*   [State Statistics Committee: fixed capital investment — 584.2 trillion soums](https://stat.uz/ru/press-tsentr/novosti-goskomstata/70884-zbekistonda-investitsiyalar-azhmi-584-2-trln-s-mga-etdi-3), 30.10.2026. The growth figure of +20.1% is calculated under the official methodology in comparable prices; the investment amount is given in current prices.
*   [State Statistics Committee: structure of investment by sector](https://stat.uz/ru/press-tsentr/novosti-goskomstata/70908-zbekistonda-ajsi-so-alarga-k-pro-investitsiya-kiritilmo-da-4), 02.11.2026.
*   [SIAT: metadata for the fixed capital investment growth indicator](https://siat.stat.uz/data/1929/?lang=en).
*   [President: enterprises not operating at full capacity](https://president.uz/en/lists/view/9140), 24.04.2026; 210 of 688 enterprises had not reached full utilisation, with forgone output estimated at 33 trillion soums.
*   [President: measures to support the sustainable operation of enterprises](https://president.uz/en/lists/view/9175), 05.05.2026; a separate estimate of additional output and exports of up to 65.6 trillion soums. Do not add this to the April estimate.
*   [NGMK: operating results for H1 2026](https://www.ngmk.uz/wp-content/uploads/2026/07/press-reliz-1yy2026-RU.pdf), 14.07.2026; physical output, investment programme, localisation, inter-sectoral cooperation and capacity expansion projects.
*   [NGMK: IFRS financial results for H1 2026](https://www.ngmk.uz/wp-content/uploads/2026/09/NMMC-FY-1H_2026-IFRS-RU_v2.pdf), 28.09.2026; revenue, EBITDA, profit, AISC, Cash CAPEX, operating cash flow and debt load.
*   [NGMK: operating results for the nine months of 2026](https://www.ngmk.uz/ru/press-center/news-and-press-releases/2026-yil-9-oylikdagi-ishlab-chiqarish-natijalari/), 21.10.2026; physical output and year-on-year comparisons of the investment programme, localisation and cooperation.

### Domestic demand and imports

*   [State Statistics Committee: imports by main commodity group, January–July 2026](https://stat.uz/en/press-center/news-of-committee/70320-zbekistonga-chet-eldan-eng-k-p-nimalar-import-ilinmo-da-5), 24.09.2026: food $3.3 billion (+41.9% in dollar terms), machinery and transport $9.7 billion (+20%).
*   [State Statistics Committee: foreign trade for January–September 2026](https://stat.uz/en/press-center/news-of-committee/70803-zbekistonning-tash-i-savdo-ajlanmasi-9-ojda-65-4-mlrd-dollarga-etdi-4), 28.10.2026: imports $39.6 billion (+15.7%), exports $25.8 billion (−5.5%), trade deficit $13.7 billion, goods excluding gold $13.3 billion (+31.6%).
*   [State Statistics Committee: agriculture for January–September 2026](https://stat.uz/en/press-center/news-of-committee/70845-agriculture-forestry-and-fisheries-in-uzbekistan-for-january-september-2026), 29.10.2026: agricultural output +4.6% in real terms; individual physical volumes moved unevenly.
*   [Retail trade for January–September 2026](https://www.uzdaily.uz/en/uzbekistans-retail-trade-turnover-rises-197-to-4322-trillion-soums/), 03.11.2026: +19.7% in real terms; source — preliminary State Statistics Committee data.
*   [Central Bank: remittances by individuals in H1 2026](https://cbu.uz/en/press_center/releases/4359624/), 14.08.2026: $9.3 billion, +13% year on year.
*   [Central Bank: loans by sector, as of August 1, 2026](https://cbu.uz/en/statistics/bankstats/4470285/): the entire portfolio +12% year on year; individuals +19%, construction +32%, trade and catering +25%, agriculture +14%, industry −13%; the loans-to-individuals category is broader than consumer lending.

### Exports and the balance of payments

*   [Export structure for January–September 2026](https://www.spot.uz/ru/2026/10/31/gold-pause): exports of non-monetary gold — $2.8 billion (10.8% of total exports), goods excluding gold — $13.3 billion, services — about $9.7 billion. This is a breakdown of the $25.8 billion total; the $13.3 billion cannot be read as all export earnings excluding gold, since services are counted separately.
*   [Central Bank: balance of payments, international investment position and external debt for H1 2026](https://cbu.uz/en/press_center/releases/4616748/), 29.09.2026. Balance of payments indicators should not be directly subtracted from the State Statistics Committee's foreign trade statistics in order to derive third-quarter values.
*   [Central Bank: Precious Metals Storage and Operations Division](https://cbu.uz/en/about/central-office/divisions/service-for-storage-and-work-with-precious-metals/), functions include arranging purchases of metals from industrial enterprises.
*   [Law "On Precious Metals and Precious Stones", Article 21](https://lex.uz/ru/docs/6118259?query=user): the state's priority right and the Central Bank's offer of refined gold.
*   [IMF: 2025 Article IV Consultation, Informational Annex](https://www.elibrary.imf.org/view/journals/002/2025/143/article-A002-en.xml): the Central Bank is described as a direct buyer of gold produced in Uzbekistan and a supplier of foreign currency to the market in an equivalent soum amount.
*   [Central Bank: Domestic Foreign Exchange Market Review for January–September 2026, Spot summary](https://www.spot.uz/en/2026/11/03/foreign-currency/), 03.11.2026: demand $50.4 billion, supply excluding Central Bank operations $43.8 billion, gap $6.6 billion versus $7.79 billion for January–September 2025; liquidity from monetary gold operations was sterilized through foreign exchange operations and other monetary policy instruments in line with the regulator's neutrality principle.
*   [Kursiv: Gold trading on the Uzbek Commodity Exchange in 9M-2026](https://uz.kursiv.media/2026-10-30/prodazha-zolota-v-uzbekistane/), 30.10.2026: 3.73 t in 9M-2025 and 8.84 t in 9M-2026; about 90% of purchases are by jewellery enterprises.
*   [World Gold Council: Central Bank Gold Statistics, June 2026](https://www.gold.org/goldhub/gold-focus/2026/08/central-bank-gold-statistics-june-2026), 04.08.2026: reported net purchases by Uzbekistan for the first half of the year — about 41 t.
*   [Interfax: Gold exports and output in Uzbekistan](https://interfax.com/newsroom/top-stories/118895/), 28.08.2026: reports gross gold exports by the Central Bank of about 100 t for 2025; this is not the net change in reserves and needs to be reconciled with the physical stock series.
*   [Interfax: Uzbekistan's Central Bank bought 109 t and exported 100 t of gold in 2025](https://www.interfax.ru/business/1102216), 09.07.2026: purchases from producers 109 t (110 t in 2024), exports 100 t (100 t in 2024), net change in the physical stock for 2025 — about +0.3 million ounces; output 118.4 t.
*   [Gazeta.uz: Uzbekistan among the three largest gold buyers](https://www.gazeta.uz/ru/2026/09/07/gold), 07.09.2026: minus 1 t in July, about 40 t since the start of the year, about 431 t at the end of July; the last gold shipment abroad was in September 2025, worth roughly $1.5 billion.
*   [World Gold Council: Central Bank statistics, July 2026](https://www.gold.org/goldhub/gold-focus/2026/09/central-bank-gold-statistics-central-banks-make-positive-headlines-gold), 03.09.2026: 1 t sold in July, about 40 t since the start of the year, stock about 431 t (87% of reserves).
*   [World Gold Council: Central Bank statistics, August 2026](https://www.gold.org/goldhub/gold-focus/2026/10/central-bank-gold-statistics-central-banks-continue-summer-spree-august), 06.10.2026: plus 8 t in August, about 50 t since the start of the year.
*   [UzDaily: Uzbekistan's Central Bank bought 8 tonnes of gold in August — WGC](https://www.uzdaily.uz/ru/tsb-uzbekistana-v-avguste-kupil-8-tonn-zolota-wgc), 07.10.2026: stock about 439 t at the end of August.
*   [GoldSilver: Gold Reserves by Country — 2026 Rankings](https://goldsilver.com/industry-news/article/gold-reserves-by-country-the-2026-rankings), 28.05.2026: plus 25 t in the first quarter of 2026, stock 416 t, 87% of reserves.
*   [NGMK: Operating results for 9M-2025](https://www.ngmk.uz/en/press-center/news-and-press-releases/2025-yil-9-oylikdagi-ishlab-chiqarish-natijalari/), 10.10.2025: 2.3466 million ounces of output.
*   [NGMK: Operating results for 9M-2026](https://www.ngmk.uz/ru/press-center/news-and-press-releases/2026-yil-9-oylikdagi-ishlab-chiqarish-natijalari/), 21.10.2026: about 2.3 million ounces; the figure is rounded.
*   [AGMK: Analytical note for 9M-2025](https://agmk.uz/storage/uploads/docs/df2820ee6406f6874383fc8751756a3d.pdf), refined gold target fulfilment — 100.7%, tonnage not disclosed.
*   [AGMK: Analytical note for 9M-2026](https://agmk.uz/storage/uploads/docs/analitika-9-oi-2026-ru-6acf749e990e2439869438_6ad0d033449390.16337612.pdf), refined gold target fulfilment — 101.3%, tonnage not disclosed.
*   [Central Bank: International reserves at the start of October 2026, data and dynamics](https://www.spot.uz/en/2026/10/09/gold-reserves/), 09.10.2026: 446.2 t of gold and $5.92 billion in the foreign currency portion of reserves as of 01.10.2026.
*   [World Bank Pink Sheet: Commodity prices](https://thedocs.worldbank.org/en/doc/74e8be41ceb20fa0da750cda2f6b9e4e-0050012026/world-bank-commodities-price-data-the-pink-sheet): monthly average gold prices, from which the average price for January–September 2025 and 2026 is calculated.

### Independent indicators, mirror statistics and estimates

*   [Ministry of Energy: Electricity generation in 9M-2026 — 69.6 billion kWh (+8.4%), consumption +6.3%, imports −27.2%](https://daryo.uz/ru/2026/10/22/v-uzbekistane-proizvodstvo-elektroenergii-viroslo-na-8-4-import-sokratilsya-na-27-2), 22.10.2026.
*   [State Statistics Committee: Transport services in 9M-2026 — 162.8 trillion soums, +16.4% in real terms](https://stat.uz/ru/press-tsentr/novosti-goskomstata/70924-zbekistonda-transport-khizmatlari-azjmi-162-8-trln-s-mga-etdi-3), 03.11.2026.
*   [APA: Passenger car production in 9M-2026 — 356.3 thousand (+9.2%), specialised vehicles — 43 thousand versus 39.1 thousand](https://www.abauzbekistan.uz/ru/news/press-relizy/proizvodstvo-avtomobilej-v-uzbekistane-vyroslo-na-9.htm), 30.10.2026.
*   [Cement production by large enterprises in 1H-2026 — 10.1 million tonnes](https://newslineuz.com/ru/article/1255583), 10.08.2026.
*   [UzDaily: Freight turnover in 1H-2026 +43%, transit +35%](https://www.uzdaily.uz/ru/gruzooborot-uzbekistana-vyros-na-43-za-pervoe-polugodie-2026-goda), 02.07.2026 — the series is inconsistent with the rest of the physical statistics and itself requires verification.
*   [Chinese Customs: Trade with Uzbekistan in January–October 2025](https://caliber.az/en/post/china-uzbekistan-trade-sees-significant-growth-in-first-ten-months-of-2025), 18.11.2025: imports from Uzbekistan $1.5 billion (−6.5%), exports $11.38 billion (+21.3%).
*   [State Statistics Committee: Trade with China in January–August 2025, Spot summary](https://www.spot.uz/ru/2025/10/01/china-trade), 01.10.2025: exports $1.24 billion (−9%), imports $8.45 billion (+27.3%), gas $494.7 million versus $563.2 million in Chinese statistics.
*   [IMF: Article IV Consultation, June 2026](https://www.imf.org/en/news/articles/2026/06/18/pr-26214-uzbekistan-imf-executive-board-concludes-2026-article-iv-consultation): Q1 growth — 8.7%, 2026 forecast — 6.8%.
*   [IMF: WEO October 2026 — Uzbekistan forecast raised to 7.8% for 2026 and 6.5% for 2027](https://www.imf.org/external/datamapper/NGDP_RPCH@WEO/UZB); [Gazeta.uz summary](https://www.gazeta.uz/en/2026/10/16/uzbekistan-gdp-imf), 16.10.2026.
*   [World Bank: Uzbekistan growth forecast for 2026 — 7.9%, the highest in the ECA region](https://caspianpost.com/uzbekistan/uzbekistan-set-for-highest-gdp-growth-in-central-asia-world-bank), October 2026.
*   [S&P: Sovereign rating upgraded from BB− to BB, outlook "stable"](https://www.spglobal.com/ratings/en/regulatory/article/-/view/type/HTML/id/3484207), 21.11.2025.
*   [World Economics: Independent estimate of growth (5.6% for 2024) and statistics quality (C)](https://www.worldeconomics.com/grossdomesticproduct/gdp-annual-growth-rate/Uzbekistan.aspx).
*   [Gazeta.uz: Central Bank statement on lowering loan and deposit rates over the next 3–4 years](https://www.gazeta.uz/ru/2026/11/04/deposits-credits), 04.11.2026; Spot, 04.11.2026: at the November meeting the Central Bank discussed cutting the policy rate to 13.5%, the decision was to keep it at 14%.
*   [Gazeta.uz: Sales of gold bars to the public](https://www.gazeta.uz/ru/2026/04/07/gold), 07.04.2026: 236 kg in 2024, 661 kg in 2025 (versus 558 kg in 2020–2024), returns of 31.6% and 54.5% versus 19.5% and 19.3% on soum time deposits.
*   [Gazeta.uz: Remittances in 9M-2026 — $15.17 billion (+8.9%), net balance $13.2 billion](https://www.gazeta.uz/ru/2026/11/03/money-transfers), 03.11.2026; [Central Bank foreign exchange market review as summarized by Spot](https://www.spot.uz/ru/2026/11/02/remittances), 02.11.2026.
*   [Gazeta.uz: Remittances in 2025 — $18.9 billion (+27.2%), share of household income 16.5% → 21%](https://www.gazeta.uz/ru/2026/01/28/money-transfers), 28.01.2026; [remittances from Russia in 2024 — $11.5 billion (+29%)](https://plusworld.ru/articles/62158), January 2025.
*   [State Statistics Committee: GDP deflator for 1H-2025 — 112.4%](https://invexi.org/ru/press/dynamics-and-structure-of-uzbekistan-s-gdp-for-the-1st-half-of-2025), July 2025.
*   [Gazeta.uz: New CPI methodology from 2026 — updated basket and formula](https://www.gazeta.uz/ru/2026/02/04/inflation), 04.02.2026.
*   Self-employed: registration according to the tax committee — [3.5 million (July 2024)](https://upl.uz/economy/43462-news.html), 20.07.2024; [over 4 million (October 2024)](https://upl.uz/economy/45898-news.html), 22.10.2024; [4.7 million (March 2025)](https://repost.uz/kogda-sam-sebe-biznesmen), 31.03.2025; labour force survey series (0.56 million in 2020; 4.47 million and 45.7% of the employed in 2024; 4.14 million in 2025) — per State Statistics Committee/SIAT publications of 2025–2026.

### Procurement, productivity and competition

*   [The Audit Chamber: results for the first half of 2026](https://ach.gov.uz/en/events-nav/news/detail/6a59d56d3a716b2dfd467e25), 16.07.2026. Violations of procurement legislation identified — 9.1 trillion soums; this is not an estimate of proven damage.
*   [The Anti-Corruption Agency: violations in public procurement in 2025](https://kun.uz/en/news/2026/02/26/anti-corruption-agency-reports-uzs-6666-billion-in-flawed-public-tenders-in-2025), 26.02.2026.
*   [OECD: productivity and the impact of foreign investment in Uzbekistan](https://www.oecd.org/en/publications/roadmap-for-sustainable-investment-policy-reforms-in-uzbekistan_20865f29-en/full-report/trends-and-impacts-of-fdi-in-uzbekistan_1794e285.html). An estimate of the productivity gap between foreign-owned companies and domestic firms; it does not isolate the effect of management from size, capital intensity and sectoral composition.

### Theory, institutions and historical comparisons

*   [SWP: Religious Policy in Uzbekistan — Between Liberalisation, State Ideology and Islamisation](https://www.swp-berlin.org/10.18449/2023RP08/), Andrea Schmitz, 21.08.2023; a study of the religious media field and online preachers. It describes the environment and the influence of religious channels but does not measure the prevalence of pseudoscientific videos.
*   [Sayfitdinova: Spiritual Educational Leaders in Uzbekistan’s Blogosphere — A Case Study Approach](https://eoconf.com/index.php/icset/article/view/437), 29.11.2025; a qualitative study of individual YouTube, Telegram and Instagram channels, not a representative survey of the population.
*   [Leonid Krutakov: "Oil and the World. The Seven Sisters — the House That John Built"](https://ripol.ru/books/26/3598/), RIPOL klassik, 2025; the author's historical and political-economic interpretation of the role of hydrocarbons in the structure of the modern economy.
*   [Bank of England: Money creation in the modern economy](https://www.bankofengland.co.uk/quarterly-bulletin/2014/q1/money-creation-in-the-modern-economy), 14.03.2014; an explanation of how deposit money is created through bank lending.
*   [Steve Keen: A monetary Minsky model of the Great Moderation and the Great Recession](https://www.sciencedirect.com/science/article/pii/S0167268111000266), *Journal of Economic Behavior & Organization*, vol. 86, 2013, pp. 221–235; endogenous money, credit, debt and financial instability in a Minsky model.
*   [Central Bank of Uzbekistan: monetary policy review for Q2 2026](https://cbu.uz/en/press_center/releases/4565703/), 22.09.2026; inflation, domestic demand and household and business expectations.
*   [Kun.uz: household and business inflation expectations in September 2026](https://kun.uz/en/news/2026/11/03/inflation-expectations-ease-among-households-and-businesses-in-uzbekistan-59143), 03.11.2026; expected household inflation — 8.7%, in Tashkent — 11.6%; perceived inflation and regional differences.
*   [UzDaily: household inflation expectations in September 2026](https://www.uzdaily.uz/en/uzbekistans-inflation-expectations-fall-to-87-in-september/), 03.11.2026; regional and social differences.
*   [Gazeta.uz: Central Bank on the gradual decline in deposit and lending rates](https://www.gazeta.uz/ru/2026/11/04/deposits-credits/), 04.11.2026.
*   [Gazeta.uz: the decision on the policy rate and inflation](https://www.gazeta.uz/ru/2026/11/05/main-rate-inflation/), 05.11.2026.
*   [Decree of the President of the Republic of Uzbekistan No. UP-43 of 05.03.2025 on the mass valuation of real estate](https://lex.uz/en/acts/7417793); its goals include establishing the value of property as an economic asset and aligning valuations for taxation, sale and purchase, collateral and other transactions.
*   [my.gov.uz: results of the mass valuation of real estate in Tashkent](https://my.gov.uz/uz/news/52dbe497-64ba-4c26-b91d-d615814f9db4), 07.09.2026; over 600 thousand apartments — 426.7 trillion soums against the previous 57.8 trillion; over 213 thousand houses — 548.4 trillion against 39.1 trillion.
*   [Central Bank: LTV ratio for mortgage loans](https://cbu.uz/ru/financial-stability/macroprudential-policy-tools/loan-to-value/), updated 10.03.2026; a general limit of 80% with limited exceptions.
*   [OECD: PISA 2022 Country Note — Uzbekistan](https://www.oecd.org/en/publications/pisa-2022-results-volume-i-and-ii-country-notes_ed6fbcc5-en/uzbekistan_2bb94bf1-en.html), 05.12.2023; 19% of students reached the baseline level in mathematics, 14% in reading and 19% in science. These are the results of schoolchildren in 2022, not a characteristic of the entire workforce.
*   [Government of Uzbekistan: measures to cut red tape](https://www.government.uz/ru/digital/news/view/164274), May 2026; 5,650 functions of state bodies, more than 42 thousand mandatory requirements for business and 1,041 types of public services.
*   [Steve Keen, Robert U. Ayres, Russell Standish: A Note on the Role of Energy in Production](https://www.sciencedirect.com/science/article/pii/S0921800917311746), *Ecological Economics*, 2019, DOI: 10.1016/j.ecolecon.2018.11.002.
*   [Hyman P. Minsky: The Financial Instability Hypothesis](https://www.levyinstitute.org/publications/the-financial-instability-hypothesis/), Working Paper No. 74, 1992.
*   [Irving Fisher: The Debt-Deflation Theory of Great Depressions](https://fraser.stlouisfed.org/title/debt-deflation-theory-great-depressions-3596/fulltext), *Econometrica*, 1933.
*   [Bank of Japan: Explaining Asset Bubbles in Japan](https://www.imes.boj.or.jp/research/abstracts/english/me14-1-6.html), 1996; the initial rise in asset prices was linked to a sharp expansion of bank lending to real estate, and rising land values strengthened companies' collateral. [A speech by the Deputy Governor of the Bank of Japan on the credit cycle and real estate values](https://www2.boj.or.jp/archive/en/announcements/press/koen_1999/ko9909a.htm), 1999.
*   [IMF: Lessons from Southeast Asia](https://www.imf.org/en/news/articles/2015/09/28/04/53/spmds9716), 13.11.1997; real estate, unhedged foreign-currency borrowing, short-term capital inflows and weak banks in Thailand before the crisis. [IMF: Article IV consultation with Thailand](https://www.imf.org/en/news/articles/2015/09/28/04/53/pn9844), 25.06.1998; the fixed exchange rate, short-term external debt, and problems in real estate and the financial sector.
*   [IMF: Construction Activity Can Signal When Credit Booms Go Wrong](https://www.imf.org/en/Blogs/Articles/2020/02/12/blog-construction-activity-can-signal-when-credit-booms-go-wrong), 12.02.2020; Spain: private credit to GDP nearly doubled in 2000–2007, construction employment rose by 47% against 27% for the economy as a whole. [BIS: The 2008–14 banking crisis in Spain](https://www.bis.org/publications/fsicrisis-management-4-2008-14-banking-crisis-spain), 06.07.2023; risk concentrated in real estate at the savings banks, weak corporate governance and unstable wholesale funding. [ESM: Why did Spain need financial assistance?](https://www.esm.europa.eu/content/why-did-spain-need-financial-assistance), 2016; credit expansion, real estate and the vulnerability of the cajas.