Free · Uzbekistan

Urgench: What a Market Price Index Leaves Out

How to read Khorezm's official price data without pretending it records a particular seller's working day

Uzbekistan Pathways | UZ-CG3-F0034 | FREE original educational article
Mapped territory: Uzbekistan — Khorezm (Xorazm) Region; Urgench as the local teaching anchor
Evidence checked: 5 November 2026
Stage: Original source draft; independent exact-source Qwen and GPT adjudication pending; not approved for publication
Mapped deliverable: Index-versus-interview evidence comparison and reusable public-data worksheet

A customer asks whether food has become more expensive. A vendor replies that the price of one item has changed, but another has not. An official monthly price index appears on a government website. All three statements may be true. They answer different questions. The customer reports a household's experience, the vendor knows transactions and supply conditions at a particular stall, and the statistician estimates a pattern across selected goods and services. Confusing these questions produces a misleading story even when the official number has been copied correctly.

This article uses Urgench, the principal city of Uzbekistan's Khorezm Region, as a place to learn how official consumer-price figures should—and should not—be interpreted. It is not an account of an observed Urgench market, a recorded vendor interview, or a survey of a particular household. The figures below come from the National Statistics Committee's public statistical tables for Xorezm Region, not from a city-level dataset or a named stall. Their geographical boundary matters. A regional index cannot by itself describe the price at a specific market in Urgench, let alone why an individual seller changed a price.

1. Locate the exact statistic before interpreting it

The National Statistics Committee makes public price-index series available through its Statistics Agency Information and Analysis System (SIAT). One series is titled “The price index for all goods and services, compared to last month for Xorezm region” with indicator code 1.11.01.0039 [S1]. Another is “The price index for socially significant types of consumer goods (services), compared to last month for Xorezm region”, indicator 1.11.01.0024 [S2]. These are not interchangeable labels. The first is organised as an all-goods-and-services regional series with numerous item categories; the second is a separately published series focused on socially significant categories.

Both pages specify “compared to last month” and express values as an index with the previous month's level represented by 100. A displayed value of 100.6 therefore describes an estimated 0.6% month-on-month increase for the defined series. It does not mean that inflation is 100.6%, or that prices are 0.6% higher than on the same date a year earlier. Equally, 99.8 describes a decline of about 0.2% against the immediately preceding month, not a reduction of 99.8% in shoppers' bills.

The public Xorezm table available during this review showed the following monthly figures for 2026, as of its October 2026 update [S1, S2]. These are selected readings, not a new index calculated by this author:

Reading rule: September's composite 100.6 means a September change relative to August, not a direct comparison with June, last September, or an individual Urgench stall. The food-products value 101.0 applies to that reported category, not to every food item. Official tables can be revised; readers should check the live indicator, publication date, definitions and export metadata rather than treating this snapshot as permanent.

The geography label here is Xorezm Region (also written Khorezm in English transliteration). The geographical anchor Urgench does not turn the table into an Urgench-only measurement. The distinction should appear in the article title, source caption and every reuse of the figure.

2. What goes into a consumer-price index?

A consumer-price index is not made by visiting every shop and averaging every price. Statistical authorities need rules for choosing representative products and services, collecting comparable prices, handling missing observations, and weighting items to reflect household expenditure. This is not arbitrary guesswork: it is a documented measurement design. But the design asks what prices are doing for the defined population and consumption basket, not what every individual person experiences.

The official methodological notes embedded in the Uzbek-language SIAT price-index publication state that from January 2026 the monthly calculation covers 517 types of goods and services, comprising 419 goods and 98 services, compared with 510 types in 2025. The notes describe price collection from the 1st through the 25th of each month, sampling of trading and service locations including stationary shops and mobile stalls, selection across regional centres and other sampled towns or district centres, and the use of household expenditure structure in forming weights [S3]. They also describe a change in the calculation formula from 2026. This is a paraphrase of the official methodological note, not a claim that 517 individual stalls were visited in Urgench.

A sample of observed outlets is not a complete census of local commerce. A representative consumption basket is not any one family's shopping list. A monthly collection window is not proof that a price was unchanged on every day outside that window. And a weighted composite is not the price of a literal package one can buy at a market. These limits matter precisely because the published figures have value: well-defined measurements should be used to answer well-defined questions.

The International Labour Organization's Consumer Price Index Manual: Concepts and Methods (2020), prepared with other international statistical organisations, describes separate technical problems of expenditure weights, sampling, price collection, product substitution, quality changes and index calculation [S4]. Those issues explain why a headline statistic cannot simply be reverse-engineered into a statement about an individual vendor's profit or a household's ability to buy dinner.

3. Indexes and prices are different kinds of numbers

Suppose a teacher presents this fictional, not locally observed, example: the price of a standard package of rice is 10,000 soums in Month A and 10,600 soums in Month B. The item's simple price relative is 106.0, a rise of 6.0%. An official food-products index could still be 101.0 over the same interval because it combines movements across many weighted products. Other components may have moved little, fallen, or carried different weights.

Conversely, a vendor could keep an advertised unit price at 10,000 soums but change the package from one kilogram to 900 grams. A price comparison ignoring the quantity would conclude there was no change. The quantity-adjusted cost per kilogram would have increased. Whether and how such a change is treated in an official price sample depends on the detailed statistical specifications; this example is an exercise in comparison discipline, not a statement about Uzbekistan's field procedure.

A seller's gross revenue is different again. Selling 25 units at 10,000 soums produces 250,000 soums in gross sales before costs. Selling 20 units at 10,600 produces 212,000 soums. A price has risen, yet gross sales are lower in this hypothetical example. The seller might also face a different wholesale purchase price, transport cost, spoilage rate, stall fee or credit arrangement. A CPI cannot reveal these details. Neither can it prove that a particular seller's income increased when the published food index rose.

Three data types must therefore be kept distinct:

• Price level: a quoted amount in soums per clearly stated unit, at a particular place and time.
• Price relative/index: the movement of a comparable item or designed basket between defined periods, with a base of 100 in the table used here.
• Income or margin: money received or retained by a trader after costs, which requires transaction and expense evidence beyond consumer-price indices.

Mixing these types is a common source of impressive-looking but empty market stories.

4. Why a regional average cannot tell the story of one Urgench household

Imagine two invented shopping lists. Household A buys more rice and vegetables; Household B spends a larger share on transport, medicine or household services. Even if both live within the coverage region, a food-products index and an all-items index need not approximate each household's personal experience equally well. The published measure is not promising to reproduce every budget. It represents a statistical population using defined scope and weights.

Now change the question from households to neighbourhoods. A family shopping at one stall near home may face a price that is higher or lower than a sampled regional price. Its preferred brand, purchase quantity, access to credit, ability to store goods or willingness to travel may differ. A very low regional average price index could coexist with a painful increase in a particular family's main purchase. Neither statement proves the other false.

This does not make the official measure irrelevant. The table can show the direction and approximate scale of price movement for its defined region and category, making comparisons possible over time. But the responsible sentence is: “The official Xorezm food-products index was 101.0 for September 2026 compared with August.” The irresponsible sentence is: “Every Urgench family's groceries became exactly 1.0% more expensive in September.” The latter silently changes the population, basket and unit of observation.

To test the lived implications of an index, a writer would need separately designed and consent-based evidence such as anonymised household expenditure diaries or structured interviews. A public government table should not be presented as if such interviews have already been conducted.

5. What a vendor could explain—and what an interview still could not prove

A vendor might be able to describe the posted price of a product, a wholesale invoice, fluctuations in supply, an order that arrived late, spoilage, or what customers chose to buy on a particular day. That is potentially valuable evidence of a local market process. But a single vendor is not a regional statistical sample. An interviewee's recollection of a past price may be honest yet approximate; a dated transaction receipt may be more precise about that one transaction, while still revealing little about the rest of the region.

A good interviewer asks: Which item, which quality, which unit, what dates, what type of transaction, and what records are available? For example, “Are you referring to wholesale cost, your retail asking price, or the price customers actually paid?” A quote without these clarifications could accidentally compare a large wholesale sack with a small retail package.

Do not invent an interview in order to humanise the data. This article uses hypothetical speakers only; it contains no testimony from named or unnamed Urgench residents. To publish a genuine vendor story, seek informed permission for interview, attribution and photographs; allow the person to decline sensitive details such as profit or supplier identity; and confirm translations of technical terms with the interviewee where possible. When permission or language comprehension is uncertain, publish a generic method instead of attributing invented experience to local people.

The limits run both ways. A vendor's account may identify a real supply interruption that official price averages do not describe. A regional price index may show a broader pattern not obvious from the vendor's own product. One does not replace the other; each helps a reader formulate questions for the other.

6. Seasonal prices make careful comparisons essential

Markets dealing with food may face seasonal availability and storage constraints. A teacher can use the public Xorezm table to examine changes in listed food categories without claiming a single cause. For example, the food-products index in the selected regional series was 99.2 for July, 100.5 for August and 101.0 for September 2026 [S1, S2]. Read correctly, the sequence records a monthly decrease followed by two monthly increases for the measured aggregate. It does not establish which harvest, transport route, import contract or weather event drove those figures.

For a simple chained interpretation, multiplying the three month-to-month relatives gives roughly 0.992 × 1.005 × 1.010 = 1.0069, or a net change of approximately +0.7% over the three transitions. This is a classroom calculation using published rounded aggregate figures; it is neither an official multi-month statistic nor a substitute for higher-precision source data. Readers should never add the listed percentages without considering compounding, or silently compare an August figure with a September value from a different index series.

A regional commodity subcategory may move in a different direction from the food aggregate. Read the item name carefully: “round-grain rice” is not simply “all rice”; “beef except boneless meat” is not “all meat.” Classification definitions protect against sweeping interpretations. When item prices are affected by quality differences, packaging and source availability, a detailed technical treatment is needed before attributing the difference to inflation alone.

The right next question after observing a change is not “What is the story?” but “What additional evidence would distinguish possible explanations?” A harvest bulletin, an official transport restriction, a transaction record or a clearly dated local survey might support a causal discussion. Without such evidence, the table shows a change, not the cause.

7. The danger of comparing unlike time periods

The same published system may contain monthly index relatives, comparisons with the corresponding month of the previous year, annual averages and cumulative figures from the start of the year. These answer different questions. A casual reader might see 100.6 in a monthly column and call it “annual inflation of 0.6%.” That is a period error. Even two correct values can be made incomparable by choosing different dates or coverage.

Before writing a comparison, make a six-field label: indicator code; geography; product category; reference period; comparison/base period; table update date. If any field is missing, find it before writing the percentage. The indicator code matters because S1 and S2 have different scopes, even when a row happens to display an identical rounded figure. The update date matters because published tables may gain new months or undergo corrections.

Two closely related claims illustrate the rule. “The Xorezm all-items monthly index for September 2026 was 100.6” is supported by S1. “The socially significant composite index in Xorezm for September 2026 was 100.6” is supported by S2. Those are independently labelled indicators. Their equality is not evidence that the composition, weights or full underlying price samples are identical.

An export from SIAT should preserve column headers, explanatory notes and the indicator metadata. A screenshot cropping out “compared to last month” should not be used to assert a trend. If a news caption needs to be short, the missing information belongs in the accompanying text, not in an undocumented assumption.

8. Why an index is not proof of a seller's motives

Suppose an article claims that sellers raised prices because tourism grew in Urgench. The regional CPI table cannot establish either the motivations of sellers or the relevant market transactions. The proper source for tourism figures would be a separately dated visitor or accommodation series; the proper source for a seller's decision would require direct evidence. Even if both tourism and some consumer prices increased, correlation would not prove that tourism caused a particular market price.

The same caution applies to claims about regulations, transport costs, currency changes, crop conditions or traders' profits. These are plausible topics to investigate but are not automatic explanations for the figures cited here. The sentence “the index rose” is a measurement claim. “The index rose because traders exploited visitors” is a causal and potentially reputational claim requiring more evidence. Reject the second until its evidence is substantially stronger.

The discipline is especially important when writing about a place that outside readers may know largely through tourism imagery. A busy market picture is not an economic census, and an abstract regional price index is not an eyewitness description. Neither should be used to stereotype local commerce.

9. A method for comparing official numbers with a possible future interview

Begin with the official source. Record exactly which product class and comparison period are covered. Then define what a local interview would need to discover that the table cannot: perhaps product-quality changes, a trader's costs, delivery schedules or customers buying in smaller quantities. Do not draft a quotation in advance as if it were already spoken.

An interview may be voluntary but not inherently representative. Note how the interviewee was chosen and whether a language intermediary was used. If a translator summarised a statement, do not substitute polished English wording in quotation marks without a verified record. Attribute a statement to the interviewee's particular experience, not to “the traders of Urgench.” Do not publish confidential business documents or identifiable customers' data without permission.

If an interview conflicts with a statistical index, resist declaring one wrong on that basis alone. A local price jump can be real while a regional basket moves less. Ask whether the item, unit, date, place and transaction type match. If they do not, the apparent contradiction may be a difference in measurement rather than a substantive disagreement.

The strongest account may say: “Official data indicate one trend in regional consumer prices, while a consenting vendor describes a different experience for a specific product in a specific period.” Such a statement would still need a real interview and its date; this article does not claim to possess one.

10. Printable worksheet: index-versus-interview evidence comparison

Use this form for any public regional price series and any independently gathered, permissioned local evidence. Leave cells blank rather than inventing missing observations.

Exercise A — decode a figure. Write the exact claim supported by Xorezm's September 2026 all-items monthly index of 100.6. The correct answer must include “Xorezm Region,” “all goods and services,” “September 2026” and “relative to August 2026.” Then write a tempting but invalid claim about “every market in Urgench” and explain why it does not follow.

Exercise B — compare unlike data. A fictional seller reports the price of a particular 900-gram package at a particular stall. The official series is a regional category index. List three pieces of information needed before treating the two observations as comparable. Product size/quality, date, and geographical or population scope are necessary starting points; business costs and transaction types may also matter.

Exercise C — test a causal story. A colleague writes: “The September increase proves that tourists made local traders raise prices.” Underline the exact point where data are turned into a causal claim. Specify what independent evidence would be required and give an appropriately cautious replacement sentence.

Exercise D — ethical reporting. Draft two questions you could ask a consenting vendor without requesting private financial information. Add the consent and translation checks you would undertake before publicly attributing an answer. If no interview takes place, mark the entire account NOT OBSERVED rather than writing a plausible anecdote.

11. A model caption that respects the evidence

Xorezm Region, September 2026. The National Statistics Committee's monthly all-goods-and-services price index (indicator 1.11.01.0039) was 100.6 relative to August 2026, equivalent to a month-on-month rise of approximately 0.6% in the defined regional series. This statistic does not establish prices or trader experiences at any individual Urgench stall. Source: SIAT, indicator page, October 2026 data update [S1].

This caption is less dramatic than “Prices soar at Urgench market,” but it is precise. It uses a real region and an observed time period without pretending that a data series contains testimony it never collected. A reader can follow the official link, inspect the date and category, and decide which additional sources would be needed for a local story.

12. What can be learned—and what remains unknown

A carefully interpreted consumer-price index helps readers ask whether prices in a defined group changed over a defined period. It can locate regional trends, support comparisons and provide a check on very broad claims. That is substantial public value, even without access to a particular stall.

What it cannot do is recover a vendor's supply invoices, a household's purchasing choices, a seller's reasons for revising prices, the exact price in one market, or the distribution of hardship across families. These questions require different sources collected with attention to sampling, permission and context. If those sources are unavailable, the responsible article says so.

Urgench is a useful anchor not because the table magically describes every street there, but because the difference between a regional statistic and a local lived experience can be taught clearly. Good locality writing does not shrink a regional average into a fictional neighbour. It shows readers exactly what the available evidence can establish, what it cannot, and how a better investigation could proceed.

Sources and use notes

[S1] National Statistics Committee of the Republic of Uzbekistan, SIAT indicator 1.11.01.0039, The price index for all goods and services, compared to last month for Xorezm region. Values consulted from the publicly displayed series through September 2026 (page updated 5 October 2026); includes indicator metadata and methodological link. https://siat.stat.uz/data/4598/?lang=en

[S2] National Statistics Committee of the Republic of Uzbekistan, SIAT indicator 1.11.01.0024, The price index for socially significant types of consumer goods (services), compared to last month for Xorezm region. Values consulted through September 2026; page update 6 October 2026. https://siat.stat.uz/data/4513/?lang=en

[S3] Same National Statistics Committee SIAT indicator in the original Uzbek interface, “Hisoblash usullari (qisqacha)” (summary methodology), accessed 5 November 2026. Explains 2026 basket sizes, monthly price-collection window, sampling, weights and changed index method. https://siat.stat.uz/data/4513/

[S4] International Labour Organization and international statistical partner organisations, Consumer Price Index Manual: Concepts and Methods, 2020. Methodological reference for expenditure weights, sampling, comparable prices, changes in product quality and index calculation; not a country-specific source for Xorezm. https://www.ilo.org/publications/consumer-price-index-manual-concepts-and-methods-2020

Attribution and rights: This is independently written educational text. Numeric excerpts identify government-published statistical indicators and give direct source URLs; no third-party article, photo or chart has been reproduced. No interviews, visits, stall prices, vendor motives, tourist effects, maps or private data are fabricated. Illustrative calculations and seller scenarios are explicitly hypothetical. Existing approved map SKU unchanged. Original reader worksheet appears above. Photo selection and any customer-facing layout remain subject to legitimate rights and final content QA; do not infer approval or publication from this draft.

Download PDF

Back to the library