Farrukh Turdikulov, Baseline Coverage or Marginal Job Creation? Reassessing the Role of Small Business in Employment in Uzbekistan, 2015–2025 (2026)
Summary
Article by Farrukh R. Turdikulov of Oriental University in Tashkent, published 30th Aug 2026 in the American Journal of Economics and Business Management, 9(8), 492-505, on small business and employment in Uzbekistan. It introduces a marginal contribution to employment growth coefficient and applies it to official statistics for 2015–2024. Small business accounted for 74.9 percent of total employment but only 34.5 percent of the employment increase over the period. The coefficient fell from 80.8 percent in 2015–2017 to −18.8 percent in 2018–2021, then recovered to 73.9 percent in 2022–2024. The author argues employment policy should target measured net job creation rather than coverage shares.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
Cite this document
“Farrukh Turdikulov, Baseline Coverage or Marginal Job Creation? Reassessing the Role of Small Business in Employment in Uzbekistan, 2015–2025 (2026).”
Original at eprints.umsida.ac.id. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/farrukh-turdikulov-baseline-coverage-or-marginal-job-creation-reassess-6b4f6c58da/
, SHA-256 fa5c40051b8654f489165fe1717c40adb7d05ba150d47b128c245da0aa39d1fb.Full text
American Journal of Economics and
Business Management
Vol. 9 Issue 8 | pp. 492-505 | ISSN: 2576-5973
Available online @ https://www.globalresearchnetwork.us/index.php/ajebm
Article
Baseline Coverage or Marginal Job Creation? Reassessing the
Role of Small Business in Employment in Uzbekistan, 2015–
2025
Farrukh R. Turdikulov*1
1. Senior Lecturer, Department of Economics, Oriental University, Tashkent, Uzbekistan
*Correspondence: turdikulovfarrukh@gmail.com
Abstract: Small business is routinely credited with carrying the labour market of Uzbekistan, since
the sector accounts for roughly three quarters of total employment. This study argues that such
Citation: Turdikulov F. R. share-based reasoning conflates two distinct properties — the stock of jobs a sector holds and its
Baseline Coverage or Marginal Job marginal capacity to generate new ones — and that doing so produces misleading policy
Creation? Reassessing the Role of
Small Business in Employment in conclusions. To separate them, the paper introduces the marginal contribution to employment
Uzbekistan, 2015–2025. American growth (MCE) coefficient, which expresses the change in sectoral employment as a proportion of
Journal of Economics and Business the change in economy-wide employment over the same interval. The coefficient is applied to
Management 2026, 9(8), 492-505.
official data of the National Statistics Committee of the Republic of Uzbekistan for 2015–2024,
Received: 19th May 2026 complemented by segmented trend estimation, a multiple regression of the sectoral employment
Revised: 20th Jun 2026
share on value-added and investment shares, and a three-scenario projection to 2030. The results
Accepted: 17th Jul 2026
Published: 30th Aug 2026 reveal a pronounced divergence. While small business held 74.9 per cent of total employment, it
accounted for only 34.5 per cent of the employment increment generated over 2015–2024 — less
than half its stock share. By stage, the coefficient fell from 80.8 per cent in 2015–2017 to −18.8 per
Copyright: © 2026 by the authors.
cent in 2018–2021, before recovering to 73.9 per cent in 2022–2024. Trend estimation confirms that
Submitted for open access the series is not homogeneous: a linear trend fitted to the full period is uninformative (R² = 0.011),
publication under the terms and whereas the same specification fitted to the post-pandemic recovery segment explains 96.3 per cent
conditions of the Creative
of the variation (R² = 0.963). The findings support reorienting employment policy from coverage
Commons Attribution (CC BY)
license targets towards measured net job creation, and offer a low-data-requirement diagnostic transferable
(https://creativecommons.org/lice to economies where firm-level job-flow statistics are unavailable.
nses/by/4.0/)
Keywords: Small business; employment growth; marginal contribution; net job creation; structural
breaks; segmented trend; labour market; Uzbekistan; transition economies.
1. Introduction
The capacity of an economy to absorb a growing labour force is a defining constraint
on development in countries with sustained demographic expansion. Uzbekistan, the most
populous state in Central Asia, adds a substantial cohort to its working-age population
each year, and the question of where those entrants will find employment has become
central to national economic policy. Within this policy debate, small business and private
entrepreneurship occupy a privileged position. The sector requires comparatively little
start-up capital, adjusts quickly to shifts in market conditions, draws on local labour and
raw materials, and can be established in peripheral regions within short time frames.
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
493
The scale of the sector in the national labour market appears to justify this
prominence. According to the National Statistics Committee of the Republic of Uzbekistan,
10,628.7 thousand persons were employed in small business entities in 2024, and the
sector's share of total employment stood at 74.9 per cent in 2025 [1]. In other words, more
than two of every three employed citizens hold a job located in this sector. The legal
recognition of self-employment, family entrepreneurship and home-based work in the
Law on Employment of the Population provided the institutional foundation for this
coverage e [2].
Policy targets have been set accordingly. Presidential Decree PF-50 of 19 March 2025
requires that the share of small and medium-sized business in national employment reach
78 per cent by 2030 and that more than two million new jobs be created within the same
horizon [3]. Meeting these targets is not a matter of preserving the sector's existing
employment base; it requires a substantial acceleration in the rate at which the sector
generates additional positions.
Yet a methodological gap persists in how the sector's labour market role is assessed.
The dominant practice — in national statistical reporting, in policy documents and in much
of the domestic literature — is to measure that role through the sector's share of total
employment. This indicator captures an accumulated, essentially static property: the stock
of jobs the sector currently holds. It says nothing about the sector's contribution to the flow
of new jobs. The two can diverge sharply. A sector may hold a large and slowly eroding
employment base while contributing only marginally to the growth of employment, and
in such a configuration a high share becomes an actively misleading signal about sectoral
dynamism.
The theoretical antecedent for viewing small firms as employment engines originates
in Schumpeter's concept of creative destruction [4], while Birch [5] empirically popularized
the proposition that small firms generate the majority of net new jobs. However,
subsequent literature qualified these assertions. Davis et al. [6] demonstrated that although
gross job creation in small firms is high, gross destruction is similarly high, rendering net
creation far more modest. Haltiwanger et al. [7] further showed that once controlling for
firm age, the size effect largely vanishes — job creation is concentrated in young firms
rather than small firms per se. Neumark et al. [8] confirmed an inverse relationship
between firm size and job growth, though notably weaker than Birch's early estimates.
In developing and transition context, Beck et al. [9] linked small business prevalence
with growth but cautioned against causal claims. Ayyagari et al. [10] found that while
small and mature firms hold the largest job stocks, job flows originate in small and young
firms. Nichter and Goldmark [11] stressed that many small enterprises operate at
subsistence scale without growth ambitions. Furthermore, Thurik et al. [12] highlighted a
two-way relationship between self-employment and unemployment, where necessity
entrepreneurship inflates coverage without expanding high-quality labour demand.
Audretsch et al. [13] linked knowledge entrepreneurship to productivity, while Storey [14]
advocated evaluating firm birth, growth, and survival systematically. In OECD economies,
SMEs generate 50–60% of value added [15]; in Uzbekistan, small business employment
shares exceed value-added shares by a much wider margin.
In domestic literature, Abdurahmonov [16] analyzed labour resource utilization,
Abdurahmonova [17] examined small business employment quality and decent work,
G'ulomov [18] highlighted organizational flexibility, Salaev [19] modeled development
trends, and Ibragimova [20] studied structural efficiency. However, aggregate national
studies retain a reliance on static coverage indicators.
This study addresses this gap in three respects:
1. It introduces the marginal contribution to employment growth (MCE) coefficient, an
aggregate diagnostic separating baseline coverage from marginal job-generating
capacity using official series.
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
494
2. It rejects the treatment of 2015–2025 as a homogeneous period, identifying three
qualitatively distinct stages.
3. It evaluates forward trajectories through segmented trend models and a three-
scenario projection to 2030.
The remainder of this paper is structured as follows. Section 2 sets out the data and
methodology. Section 3 presents empirical results. Section 4 discusses policy
interpretations, Section 5 draws conclusions, and Section 6 outlines limitations and future
directions.
2. Data and Methodology
2.1 Data sources and coverage
The empirical base consists of official statistical series for 2015–2025 published by the
National Statistics Committee of the Republic of Uzbekistan through its SIAT database [1],
supplemented by data from the Central Bank of Uzbekistan [21] and the Ministry of
Economy and Finance [22], and by publications of the OECD, the International Labour
Organization, the United Nations Development Programme and the International
Monetary Fund [15], [23]–[25].
Three features of the data require explicit statement before the analysis proceeds.
First, the series on active small business entities is compiled as of 1 January of each year
and excludes dehkan and farming enterprises. Second, state registers were harmonised
from 2022 onwards, and the resulting decline in recorded entity numbers over 2023–2025
is therefore partly a statistical artefact rather than an economic contraction; it should not
be interpreted as evidence of sectoral decline. Third, the continuous official series for the
absolute number of persons employed in small business extends to 2024, while for 2025
only the share indicator has been released. All computations involving absolute
employment levels are consequently restricted to 2015–2024, and 2025 values are used only
where share indicators are concerned.
The category 'small business' follows the national statistical definition, which
encompasses microfirms, small enterprises, individual entrepreneurs, family businesses,
craftsmen and, in the aggregate employment series, dehkan farms. This is broader than the
conventional OECD definition of small and medium-sized enterprises, and cross-country
comparisons in this paper should be read with that difference in mind.
2.2 Reconstruction of economy-wide employment
The indicator developed below requires a series for economy-wide employment.
Rather than drawing this series from a separate publication with potentially different
coverage conventions, it is reconstructed from the two small business series themselves,
which guarantees internal consistency:
Eʲᵃᵐⁱₜ = Eᵏᵇₜ / sₜ , (1)
where E ʲᵃᵐⁱ denotes total employment in the economy, E ᵏᵇ denotes employment in
small business entities, and s is the sector's share of total employment expressed as a
proportion. Because both inputs originate in the same statistical publication and are
compiled under identical coverage rules, the reconstructed series is free of the coverage
mismatch that would arise from combining sources. The reconstructed values are reported
in Section 4 and are consistent in magnitude with independent estimates of the Uzbek
labour force, though the reader should note that they inherit any measurement error
present in the underlying share series.
2.3 The marginal contribution to employment growth (MCE) coefficient
The central methodological contribution of this paper is the marginal contribution to
employment growth (MCE) coefficient. Its logic can be stated simply. The employment
share measures what proportion of existing jobs a sector holds; the MCE coefficient
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
495
measures what proportion of newly created jobs a sector supplies. Formally, for an interval
running from t₀ to t₁:
MCE = (Eᵏᵇₜ₁ − Eᵏᵇₜ₀) / (Eʲᵃᵐⁱₜ₁ − Eʲᵃᵐⁱₜ₀) × 100 % , (2)
where the numerator is the change in sectoral employment and the denominator is
the change in economy-wide employment over the same interval, both expressed in
thousand persons.
The interpretation is anchored on the sector's employment share. An MCE value
above the share indicates that the sector is functioning as the driver of employment
growth, expanding faster than the economy around it. A value approximately equal to the
share indicates that the sector is reproducing its existing position. A value below the share
indicates that employment growth is being generated predominantly elsewhere, and that
the sector's high coverage reflects inherited stock rather than current dynamism. A
negative value indicates net job destruction within the sector, meaning that aggregate
employment grew in spite of the sector rather than because of it.
Two properties of the coefficient require care in application. First, it is a ratio of
differences and therefore becomes unstable when the denominator approaches zero: in
periods of near-stagnant aggregate employment, small absolute changes in the numerator
produce large swings in the computed value. For this reason the coefficient is reported
here over multi-year intervals rather than year on year, and the underlying absolute
changes are reported alongside it so that readers can assess the stability of each estimate
directly. Second, the coefficient is a decomposition of net change and does not identify
gross flows; it cannot distinguish a sector with low creation and low destruction from one
with high creation and high destruction. It is thus a diagnostic instrument rather than a
substitute for firm-level job-flow analysis.
2.4 Trend specification and structural breaks
To characterise the temporal behaviour of sectoral employment, a linear trend was
estimated by ordinary least squares:
Eₜ = a + bt + εₜ , (3)
where Eₜ is employment in small business entities in thousand persons, t is a time
index, a is the intercept, b is the average annual change, and εₜ is a stochastic error term.
The specification was estimated first over the full period 2015–2024 and subsequently over
the post-pandemic segment 2020–2024, with the comparison of goodness-of-fit statistics
used as evidence on the presence of structural breaks. The segmentation is motivated ex
ante by two identifiable events — the statistical and economic adjustments of 2018 and the
pandemic shock of 2020 — rather than being selected to maximise fit.
2.5 Regression specification
To examine the association between the sector's labour market position and its
economic position, a multiple regression was specified:
EMPₜ = β₀ + β₁GDPₜ + β₂INVₜ + εₜ , (4)
where EMPₜ is the sector's share of total employment, GDPₜ its share of gross
domestic product, and INVₜ its share of investment in fixed capital, all expressed in
percentages.
Given only ten annual observations and the presence of common time trends in all
three series, this regression is not interpreted causally. It is used as a diagnostic device to
establish whether the sector's employment position moves with its value-creation position,
and the reported coefficients should be read as conditional associations rather than as
structural parameters.
2.6 Scenario design
For the projection exercise, a single point forecast was rejected in favour of three
alternative scenarios — pessimistic, baseline and optimistic — anchored on the official
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
496
2024 value of 10,628.7 thousand persons. The pessimistic scenario applies the long-run
average growth rate observed over 2015–2024 (approximately 0.5 per cent annually) and
assumes a simultaneous intensification of external and domestic risks. The baseline
scenario adopts the average annual increment of 181.9 thousand persons estimated from
the 2020–2024 recovery trend and assumes that the bulk of current policy is implemented
as planned. The optimistic scenario assumes a high degree of strategic target fulfilment,
requiring average annual growth of 2.74 per cent over 2024–2030.
3. Results
3.1 Development of the small business sector, 2015–2025
Table 1 reports the principal indicators of small business development in Uzbekistan
for 2015–2025.
Table 1. Key indicators of small business development in Uzbekistan, 2015–2025.
Share of total Newly
Share of Active Share of Share of
Year employment, established,
GDP, % entities, units exports, % investment, %
% units
2015 64.6 77.9 196,014 26,896 27.0 40.3
2016 66.8 78.2 207,104 31,279 26.0 40.9
2017 67.4 78.0 210,594 38,167 22.0 38.1
2018 63.5 76.3 229,666 48,922 27.2 38.0
2019 58.1 76.2 262,930 92,874 27.0 44.3
2020 56.7 74.5 334,767 93,214 20.5 46.0
2021 56.3 74.5 411,203 98,886 20.0 45.6
2022 54.3 73.9 462,834 90,177 31.2 47.9
2023 53.8 74.0 523,556 86,030 27.9 49.6
2024 53.5 73.9 417,080 77,046 34.1 52.4
2025 52.2 74.9 358,116 82,386 36.6 61.4
Note. Active entities are recorded as of 1 January of each year and exclude dehkan
and farming enterprises. The decline in entity numbers from 2024 reflects the
harmonisation of state registers introduced in 2022 and should not be read as economic
contraction. Source: compiled by the author from the SIAT database of the National
Statistics Committee of the Republic of Uzbekistan [26].
The data reveal two countervailing tendencies. The first is an erosion of the sector's
position in value creation: its share of gross domestic product fell from a peak of 67.4 per
cent in 2017 to 52.2 per cent in 2025, a decline of 15.2 percentage points. The second is the
stability of its labour market position: over the same interval the share of total employment
fell from 78.0 to 74.9 per cent, a decline of only 3.1 percentage points. Superimposed on
both is a rapid expansion of investment and export activity, with the sector's share of fixed
capital investment rising from 38.1 to 61.4 per cent and its share of exports from 22.0 to
36.6 per cent.
Read jointly, these movements indicate that the volume of capital directed into the
sector has grown rapidly without a proportional expansion of its employment position. A
substantial part of new investment appears to be flowing towards technological renewal
and capital-intensive activity rather than towards the expansion of labour demand.
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
497
Business demography reinforces this reading. The number of active entities rose
from 196.0 thousand in 2015 to 523.6 thousand in 2023, a 2.7-fold increase, before falling to
358.1 thousand by 2025 for the register-harmonisation reasons noted in Section 3.1.
Including dehkan farms, individual entrepreneurs and craftsmen, the total number of
operating small business entities reached 1,212.0 thousand as of 1 October 2025, or 32.1
units per thousand population (National Statistics Committee, 2025a).
The decisive observation is that this 2.7-fold expansion in entity numbers did not
generate a comparable expansion of employment. Employment in the sector rose from
10,170.4 to 10,628.7 thousand persons over 2015–2024, an increase of 4.5 per cent,
corresponding to average annual growth of 0.49 per cent. The disproportion suggests that
the majority of newly registered entities operate at micro scale or in self-employment
format, each generating on average very few positions — a pattern consistent with the
subsistence-scale entry documented for developing economies by Nichter and Goldmark
(2009).
3.2 Employment dynamics and the identification of three development stages
Placing sectoral employment against economy-wide employment is the next
analytical step. Using the reconstruction described in Section 3.2, total employment in the
economy rose from 13,055.7 thousand persons in 2015 to 14,382.5 thousand in 2024, an
increase of 1,326.8 thousand persons or 10.2 per cent. Over the same interval, employment
in small business rose by only 458.3 thousand persons, or 4.5 per cent. Figure 1 presents
the two series.
The figure makes clear that 2015–2024 cannot be treated as a homogeneous period.
Three qualitatively distinct stages are identifiable. Stage I (2015–2017) is characterised by
simultaneous expansion in both value creation and employment: the sector's share of gross
domestic product rose from 64.6 to 67.4 per cent and employment increased by 371.1
thousand persons. Stage II (2018–2021) is a period of structural disruption: the statistical
and economic adjustments of 2018, followed by the pandemic shock of 2020, reduced
sectoral employment from 10,318.9 thousand persons in 2019 to 9,865.7 thousand in 2020,
a contraction of 453.2 thousand persons or 4.4 per cent. Stage III (2022–2024) is a period of
recovery accompanied by capital deepening, during which employment rose by 497.6
thousand persons to reach its historical maximum.
Figure 1. Employment in small business and total employment in the economy of
Uzbekistan, 2015–2024, with the three identified development stages.
Note. Total employment is reconstructed according to equation (1). Source: author's
calculations based on data of the National Statistics Committee of the Republic of
Uzbekistan.
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
498
3.3 Baseline coverage versus marginal contribution
Table 2 reports the MCE coefficient computed according to equation (2) for each
stage and for the full period.
Table 2. Marginal contribution of small business to employment growth by stage,
2015–2024.
Change in small
Change in total Contribution of
business MCE
Stage Period employment, other sectors,
employment, coefficient, %
thousand persons thousand persons
thousand persons
Stage I 2015–2017 +371.1 +459.0 +87.9 80.8
Stage II 2018–2021 −48.2 +256.0 +304.2 −18.8
Stage III 2022–2024 +497.6 +673.3 +175.7 73.9
Full period 2015–2024 +458.3 +1,326.8 +868.5 34.5
Note. The MCE coefficient is computed according to equation (2). Total employment
is reconstructed according to equation (1). The Stage II estimate rests on a comparatively
small denominator and should be read as indicating sign and order of magnitude. Source:
author's calculations.
The results establish a pronounced divergence between the sector's baseline and
marginal positions. For the full period the coefficient equals 34.5 per cent. That is, while
small business held between 74 and 78 per cent of total employment throughout, it
supplied only one of every three additional jobs generated in the economy over 2015–2024.
Its marginal performance is less than half its baseline coverage.
Disaggregation by stage shows that this aggregate figure conceals sharply divergent
regimes. In Stage I the coefficient reached 80.8 per cent, closely matching the sector's
baseline share and confirming that small business was then functioning as the principal
driver of employment growth. In Stage II the coefficient turned negative at −18.8 per cent:
sectoral employment contracted by 48.2 thousand persons while economy-wide
employment expanded by 256.0 thousand. Employment growth during this stage was
therefore generated entirely outside small business — in large enterprises, the state sector
and budget-funded services. In Stage III the coefficient recovered to 73.9 per cent, although
the contribution of other sectors remained substantial at 175.7 thousand persons.
The magnitudes of the underlying absolute changes, reported alongside the
coefficient in Table 2, indicate that the estimates for Stages I and III are computed over
denominators of 459.0 and 673.3 thousand persons respectively and are therefore
reasonably stable. The Stage II estimate rests on a smaller denominator of 256.0 thousand
persons and should be read as indicating the sign and approximate order of magnitude
rather than a precise value. Figure 2 presents the decomposition graphically.
The scientific significance of this result lies in the separation it achieves. A high
baseline share reflects an inertial position sustained by the sheer size of the existing job
stock, and such a position can persist for extended periods even as the sector's dynamism
declines. The marginal contribution measures the capacity to generate new labour
demand, and it is this property that determines whether employment policy directed at
the sector will succeed in expanding aggregate employment.
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
499
Figure 2. Decomposition of employment growth between small business and other
sectors, and the MCE coefficient by stage.
Source: author's calculations.
3.4 Trend segmentation and structural breaks
The temporal behaviour of sectoral employment was examined through the trend
specification in equation (3). Estimated over the full period 2015–2024, ordinary least
squares yields Êₜ = 10,228.05 + 7.91t with a coefficient of determination of only R² = 0.011.
Time therefore explains approximately 1.1 per cent of the variation in employment, and a
linear trend fitted to the full period is statistically uninformative for forecasting purposes.
This result is not an anomaly to be discarded; it is the direct statistical expression of the
structural breaks identified in Section 4.2.
Restricting estimation to the recovery segment transforms the picture. For 2020–2024
the estimated model is Êₜ = 9,852.14 + 181.87t with R² = 0.963, so that 96.3 per cent of the
variation in employment during the recovery is explained by the time trend. The model
records a stable average increment of 181.9 thousand persons annually. With only five
observations, this specification should be treated as an indicative description of the
recovery trajectory rather than as a deterministic forecasting model.
The comparison of the two estimates yields a methodological conclusion of practical
importance: ignoring structural breaks when projecting small business employment
introduces systematic bias, because a trend fitted to the full period substantially
understates the sector's realised growth capacity and would lead to the setting of
unwarrantedly low policy targets.
3.5 Regression results
The regression specified in equation (4) was estimated on 2015–2024 data, yielding
EMP = 44.618 + 0.419·GDP + 0.141·INV, with R² = 0.939 and adjusted R² = 0.922. The
coefficient on the value-added share is 0.419 and is statistically significant (p ≈ 0.001):
holding other conditions constant, a one percentage point higher share of gross domestic
product is associated with an employment share higher by an average of 0.42 percentage
points. The coefficient on the investment share is 0.141, positive in direction but of lower
statistical significance (p ≈ 0.156), consistent with the effect of investment on employment
materialising with a lag.
Subject to the interpretive limits stated in Section 3.5, this result corroborates the
MCE finding through an independent route. If the sector's employment position is tightly
coupled to its value-creation position, then the sustained decline in the latter can be
expected to exert downward pressure on the former. The employment share has so far
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
500
proved far more resilient than the value-added share, but the regression suggests that this
resilience is unlikely to be indefinite.
3.6 Sectoral composition and the quality dimension of employment
The sectoral composition of small business clarifies the economic mechanism behind
these results. For January–September 2025, small business entities generated 638,221.7
billion soums of value added, equivalent to 51.5 per cent of gross value added in the
economy. The sector's share by branch was 96.9 per cent in agriculture, forestry and
fishing, 77.6 per cent in construction, 45.1 per cent in services and 26.4 per cent in industry
[27]. It further accounts for 81.8 per cent of retail turnover, 95.4 per cent of passenger
turnover and 77.0 per cent of freight turnover in road transport.
This distribution shows that small business occupies a system-forming position in
low-capital-intensity, high-labour-intensity branches while participating only marginally
in technologically advanced manufacturing. That composition explains the low MCE value
directly: the sector's employment base in traditional branches is close to saturation,
whereas a substantial part of new job creation is occurring in capital-intensive branches in
which the sector is weakly represented.
The financing environment has expanded considerably. Central Bank data indicate
that during 2025 the total stock of bank credit rose from 533.1 to 604.0 trillion soums (13
per cent), microloans from 42.4 to 48.9 trillion soums (15 per cent), and credit for
entrepreneurship development from 19.5 to 40.5 trillion soums, more than doubling.
According to the Ministry of Economy and Finance, the Entrepreneurship Development
Company had extended 19.516 trillion soums of financial support to 59,643 projects of
64,001 entrepreneurs by 1 July 2025, with an associated outcome of approximately 400
thousand jobs created or preserved.
That reporting convention is itself instructive. Because created and preserved jobs
are aggregated into a single figure, the reported outcome cannot be decomposed into net
creation and retention, and the effectiveness of the underlying expenditure cannot be
established from it. This is precisely the informational deficiency that a net-increment
measure such as the MCE coefficient is designed to remedy.
Quantitative employment indicators do not capture the quality dimension. National
Statistics Committee estimates place the gross value added of the non-observed economy
at 33.3 per cent of gross domestic product for January–September 2025, of which 25.6
percentage points are attributable to household and individual activity and 7.7 percentage
points to the hidden economy. The International Labour Organization estimates that
informal employment covers approximately 40 per cent of the Uzbek labour force
(International Labour Organization, n.d.), and a joint study by UNDP, the Central Bank
and the Centre for Economic Research and Reforms finds informality to be especially
prevalent in agriculture, construction, trade and services — precisely the branches in
which small business predominates. A share of the sector's measured employment
advantage therefore reflects the incomplete formalisation of labour relations rather than
superior job-generating performance.
3.7 Scenario projections to 2030
The projection exercise described in Section 3.6 produces the trajectories reported in
Table 3 [28]. The macroeconomic setting was specified with reference to current official
forecasts. The Central Bank projects economic growth of 7.5–8 per cent and inflation of
approximately 6.5 per cent for 2026, assessing the potential growth rate of the economy at
6–7 per cent. The International Monetary Fund's 2026 assessment is more conservative,
projecting real gross domestic product growth of 6.8 per cent in 2026 and approximately 6
per cent in 2027 . On the institutional side, the expansion of resources allocated to small
and medium-sized business financing for 2026–2027 strengthens the financial basis of the
projection.
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
501
Table 3. Scenario projections of employment in small business entities, 2025–2030
(thousand persons).
Year Pessimistic scenario Baseline scenario Optimistic scenario
2024 (base) 10,628.7 10,628.7 10,628.7
2025 10,681.8 10,761.5 10,919.9
2026 10,735.3 10,943.4 11,219.1
2027 10,788.9 11,125.2 11,526.4
2028 10,842.9 11,307.1 11,842.2
2029 10,897.1 11,489.0 12,166.7
2030 10,951.6 11,670.8 12,500.0
Change, 2024–2030 +322.9 +1,042.1 +1,871.3
Average annual rate ≈0.5 % ≈1.6–1.7 % ≈2.74 %
Note. The 2024 value is the official figure of the National Statistics Committee.
Scenario assumptions are described in Section 3.6. Source: author's projections.
Three conclusions follow from the projections. First, the sheer size of the existing
employment base means that even modest annual growth rates translate into hundreds of
thousands of new or formalised positions. Second, continuation of the recovery trend
would yield an employment base of 11.67 million persons by 2030. Third, the optimistic
outcome of 12.5 million cannot be achieved through credit expansion alone; it requires the
simultaneous effective operation of financial, export, technology, workforce development
and tax policy. Figure 3 presents the scenario corridor.
The width of that corridor is itself a finding. In 2025 the spread between scenarios is
comparatively narrow, but by 2030 it reaches 1.55 million persons. The magnitude of this
spread indicates that the employment outcome is more sensitive to the quality of domestic
economic policy than to external conjuncture.
Figure 3. Actual employment in small business, 2015–2024, and scenario projections
to 2030.
Source: author's calculations and projections.
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
502
4. Discussion
The central empirical finding of this study is the divergence between the baseline
and marginal measures of the sector's employment role: 74.9 per cent coverage against a
34.5 per cent contribution to employment growth. Three interpretations of this gap merit
consideration, and they are not mutually exclusive.
The first is structural saturation. The sector's employment base is concentrated in
trade, agriculture and construction, branches in which labour absorption is close to its
practical ceiling given prevailing technology and demand. Additional labour demand in
the economy is emerging predominantly in capital-intensive manufacturing and in
modern services, where small business participation is limited to 26.4 and 45.1 per cent
respectively. On this reading, the low MCE value reflects the sectoral composition of the
economy rather than any deficiency in small firms as such.
The second is the composition of entry. The 2.7-fold increase in registered entities
against a 4.5 per cent increase in employment implies that the average entrant generates a
very small number of positions. This is consistent with the necessity-entrepreneurship
mechanism described by Thurik et al. and with the subsistence-scale entry documented by
Nichter and Goldmark: registration statistics improve while labour demand does not
expand proportionally.
The third is measurement. Given informal employment of approximately 40 per cent
of the labour force and a non-observed economy of 33.3 per cent of gross domestic product,
part of the recorded change in sectoral employment may reflect movement across the
formality boundary rather than genuine job creation or destruction. This consideration
cuts in both directions and cannot be resolved with published aggregate data.
The negative Stage II coefficient warrants separate comment. It should not be read
as evidence that small business performed badly in absolute terms during 2018–2021; the
contraction of 48.2 thousand persons is small relative to a base exceeding ten million. Its
significance is comparative: during a period of external shock, the sector proved less able
to sustain employment than large enterprises and the public sector, which together added
304.2 thousand positions. This asymmetry in shock absorption is a policy-relevant
property that the employment share alone renders invisible.
The findings should be positioned carefully against the international literature. This
study does not claim to measure gross job flows, and it therefore neither confirms nor
contradicts the Davis et al. (1996) result on creation and destruction rates. What it does
show is that the analytical distinction those authors established on microdata leaves a
detectable signature in aggregate data, and that this signature can be extracted with a
straightforward decomposition. For economies in which linked firm-level registers are
unavailable — a description that fits much of Central Asia and many developing countries
— this offers a practicable first-order diagnostic.
5. Conclusions and Policy Implications
This study set out to distinguish the baseline employment coverage of small business
in Uzbekistan from its marginal contribution to employment growth, and to identify the
stages of the sector's development over 2015–2025. Four conclusions follow.
First, the sector's baseline labour market position remains stable and high: its share
of total employment held between 74 and 78 per cent over the period, with more than 10.6
million persons employed. This stability coexists with a 15.2 percentage point decline in
the value-added share and with a 2.7-fold increase in entity numbers that produced
employment growth of only 4.5 per cent. The pattern indicates extensive rather than
intensive sectoral development.
Second, the marginal contribution to employment growth coefficient quantifies the
divergence between baseline and marginal positions. Small business, holding 74.9 per cent
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
503
of employment, supplied only 34.5 per cent of the employment increment generated over
2015–2024. Across stages the coefficient moved from 80.8 per cent to −18.8 per cent before
recovering to 73.9 per cent.
Third, the 2015–2025 period is not a homogeneous series. Three qualitatively distinct
stages — expansion, structural disruption and recovery — are identifiable, and this is
confirmed statistically: a linear trend fitted to the full period is uninformative (R² = 0.011),
whereas the recovery-segment trend has high explanatory power (R² = 0.963).
Fourth, the sector's employment share is closely associated with its value-added
share (β₁ = 0.419; p ≈ 0.001), implying that the continuing decline in value creation is likely
to exert downward pressure on the employment share over the medium term.
Four policy implications follow. The first concerns measurement: the effectiveness of
state support programmes should be assessed against net-increment indicators rather than
coverage indicators. Present reporting practice aggregates created and preserved jobs into
a single figure; separating these outcomes statistically and incorporating a net-increment
measure such as the MCE coefficient into official monitoring frameworks would permit a
substantially more accurate assessment of programme performance.
The second concerns targeting. Because the sector's employment base in traditional
branches is close to saturation, directing incremental resources towards manufacturing,
information and communication, and professional services — branches in which small
business is currently weakly represented — would raise marginal effectiveness more than
further expansion in branches already dominated by the sector.
The third concerns forecasting practice. Trends estimated over periods containing
structural breaks understate the sector's realised growth capacity, creating a risk that
policy targets are set at unwarrantedly low levels. Segmented estimation should be
standard practice where identifiable breaks are present.
The fourth concerns the criteria of employment policy itself. Monitoring the
formality rate of jobs and value added per employed person alongside job counts would
support the sector's transition from extensive expansion to intensive development, and
would align national practice with the quality dimension emphasised in the decent work
literature.
The methodological contribution of this paper extends beyond the Uzbek case. The
MCE coefficient requires only two published series — sectoral employment and the
sectoral employment share — and is therefore computable for any sector in any economy
that publishes these aggregates. It offers a low-cost route to a distinction that has hitherto
required firm-level longitudinal data.
The analytical findings should be interpreted in light of several limitations, including
the reliance on reconstructed aggregate series, the absence of firm-level gross job flow
microdata, the constraints of short time series, and statistical disruptions caused by register
harmonization. To address these constraints, future research should:Extend the analysis
to regional panel datasets across Uzbekistan's administrative divisions. Utilize linked
administrative microdata to validate aggregate decompositions against measured gross
job flows when accessible. Perform comparative cross-country evaluations across Central
Asian economies experiencing similar data constraints. DeclarationsFunding: This
research received no specific grant from any funding agency in the public, commercial, or
not-for-profit sectors. Conflicts of Interest: The author declares no conflict of interest. Data
Availability: All data analyzed in this study are publicly available from the official
statistical sources cited in the references.
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
504
REFERENCES
[1] National Statistics Committee of the Republic of Uzbekistan, "Number of persons employed in small business
entities and its share in total employment," SIAT Database, 2026. [Online]. Available:
https://siat.stat.uz/data/425/
[2] Republic of Uzbekistan, Law No. ZRU-642 on Employment of the Population, Oct. 20, 2020. [Online]. Available:
https://lex.uz/docs/-5055690
[3] President of the Republic of Uzbekistan, Decree No. PF-50 on measures to increase the role of small and medium-
sized business in the economy, Mar. 19, 2025. [Online]. Available: https://lex.uz/docs/-7444738
[4] J. A. Schumpeter, The Theory of Economic Development. Cambridge, MA, USA: Harvard University Press, 1934.
[5] D. L. Birch, The Job Generation Process, MIT Program on Neighborhood and Regional Change, Cambridge, MA,
USA, 1979.
[6] S. J. Davis, J. Haltiwanger, and S. Schuh, "Small business and job creation: Dissecting the myth and reassessing
the facts," Small Business Economics, vol. 8, no. 4, pp. 297–315, 1996. https://doi.org/10.1007/BF00393278
[7] J. Haltiwanger, R. S. Jarmin, and J. Miranda, "Who creates jobs? Small versus large versus young," Review of
Economics and Statistics, vol. 95, no. 2, pp. 347–361, 2013. https://doi.org/10.1162/REST_a_00288
[8] D. Neumark, B. Wall, and J. Zhang, "Do small businesses create more jobs? New evidence for the United States
from the National Establishment Time Series," Review of Economics and Statistics, vol. 93, no. 1, pp. 16–29, 2011.
https://doi.org/10.1162/REST_a_00060
[9] T. Beck, A. Demirgüç-Kunt, and R. Levine, "SMEs, growth, and poverty: Cross-country evidence," NBER
Working Paper No. 11224, 2005. https://doi.org/10.3386/w11224
[10] M. Ayyagari, A. Demirgüç-Kunt, and V. Maksimovic, "Who creates jobs in developing countries?" Small
Business Economics, vol. 43, no. 1, pp. 75–99, 2014. https://doi.org/10.1007/s11187-014-9549-5
[11] S. Nichter and L. Goldmark, "Small firm growth in developing countries," World Development, vol. 37, no. 9,
pp. 1453–1464, 2009. https://doi.org/10.1016/j.worlddev.2009.01.013
[12] A. R. Thurik, M. A. Carree, A. van Stel, and D. B. Audretsch, "Does self-employment reduce unemployment?"
Journal of Business Venturing, vol. 23, no. 6, pp. 673–686, 2008. https://doi.org/10.1016/j.jbusvent.2008.01.007
[13] D. B. Audretsch, M. C. Keilbach, and E. E. Lehmann, Entrepreneurship and Economic Growth. Oxford, UK:
Oxford University Press, 2006. https://doi.org/10.1093/acprof:oso/9780195183511.001.0001
[14] D. J. Storey, Understanding the Small Business Sector. London, UK: Routledge, 1994.
[15] OECD, SMEs and Entrepreneurship, Paris, France: OECD Publishing, 2025. [Online]. Available:
https://www.oecd.org/en/topics/smes-and-entrepreneurship.html
[16] Q. X. Abdurahmonov, Mehnat iqtisodiyoti (nazariya va amaliyot) [Labour Economics: Theory and Practice].
Tashkent, Uzbekistan: Fan, 2009.
[17] G. Q. Abdurahmonova, Kichik biznesda aholini ish bilan ta'minlash [Employment of the Population in Small
Business]. Tashkent, Uzbekistan: Iqtisodiyot, 2014.
[18] S. S. G'ulomov, Tadbirkorlik va kichik biznes [Entrepreneurship and Small Business]. Tashkent, Uzbekistan:
Sharq, 1998.
[19] S. K. Salaev, "Kichik biznes rivojlanish tendensiyalarini modellashtirish va bashoratlash" [Modelling and
forecasting small business development trends], Iqtisodiyot va Moliya, no. 4, pp. 42–51, 2019.
[20] M. M. Ibragimova, "Kichik biznes va xususiy tadbirkorlik faoliyatining tarkibiy o'zgarishlar asosida
samaradorligini oshirish," PhD dissertation abstract, Tashkent, Uzbekistan, 2018.
[21] Central Bank of Uzbekistan, "Banking system statistics and monetary policy press releases," 2026. [Online].
Available: https://cbu.uz/en/statistics/bankstats/3326290/
[22] Ministry of Economy and Finance of the Republic of Uzbekistan, "Information on the activities of the
Entrepreneurship Development Company," 2025. [Online]. Available:
https://gov.uz/oz/imv/news/view/70217
[23] International Labour Organization, "Transition from informal to formal employment project, Uzbekistan,"
n.d. [Online]. Available: https://www.ilo.org/projects-and-partnerships/projects/transition-informal-formal-
employment-project-uzbekistan
[24] International Monetary Fund, "Uzbekistan: 2026 Article IV consultation," IMF Country Report, 2026. [Online].
Available: https://www.imf.org/en/news/articles/2026/06/18/pr-26214-uzbekistan-imf-executive-board-
concludes-2026-article-iv-consultation
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
505
[25] UNDP, Central Bank of Uzbekistan, and Centre for Economic Research and Reforms, Informal Employment in
Uzbekistan, Tashkent: UNDP, 2025. [Online]. Available:
https://www.undp.org/uzbekistan/publications/informal-employment-uzbekistan
[26] National Statistics Committee of the Republic of Uzbekistan, "Main indicators of small business entities in
the Republic of Uzbekistan," Press Release, Oct. 24, 2025. [Online]. Available: https://stat.uz/img/press-
relizlar/kb_uz_p46946.pdf
[27] National Statistics Committee of the Republic of Uzbekistan, "Gross domestic product of the Republic of
Uzbekistan," Press Release, Oct. 28, 2025. [Online]. Available: https://stat.uz/img/press-relizlar/analitika-
vvp_zb-lotin_p95454.pdf
[28] President of the Republic of Uzbekistan, Decree No. PF-138, Aug. 19, 2025. [Online]. Available:
https://lex.uz/docs/-7696559
American Journal of Economics and Business Management 2026, 9(8), 492-505 https://globalresearchnetwork.us/index.php/ajebm
File and source
- File
- turdikulov-2026-baseline-coverage-or-marginal-job-creation-reassessing-the-role-of-small-b.pdf
- Size
- 611,588 bytes
- SHA-256
- fa5c40051b8654f489165fe1717c40adb7d05ba150d47b128c245da0aa39d1fb
- Our copy
- turdikulov-2026-baseline-coverage-or-marginal-job-creation-reassessing-the-role-of-small-b.pdf
- Original
- eprints.umsida.ac.id