MMESP · Overview
Standalone
Contour lines of the Mongolian steppe flowing into economic model curves

One economy, eight lenses.

MMESP integrates eight economic models calibrated to Mongolia — from Leontief input–output tables to New Keynesian DSGE — so a single shock can be traced from a mine shaft to a household budget.

Models 8 Sectors 21 Horizon 20 yrs Base year 2024 Version 5.0
A model is a lens, not an oracle. MMESP looks through eight of them at once — because no single equation can hold a whole economy.
Live terrain · the steppe as a surface — drag to rotate
§1

Macro snapshot — Mongolia 2024

GDP
80.2 T MNT
Real growth
4.8 %
Mining share
25.6 %
Debt / GDP
44.5 %
Inflation
6.2 %
HDI
0.741
Population
3.42 M
Urbanization
69.2 %
§2

How a shock travels

Input–Output
Sectoral spillovers
→
CGE
Prices & factors
→
DSGE
Monetary response
→
Macro-Fiscal
Budget & debt path
SAM
Who gains, who loses
DSA
Debt stress tests
SDG
Development outcomes

↳ downstream: distribution, risk and development impacts fan out from the fiscal path

§3

Model atlas

§4

Data provenance

SourceData typeVintage
National Statistics Office (NSO)I-O table, national accounts, trade2019–2024
Bank of MongoliaMonetary policy, inflation, interest rates2020–2024
Ministry of FinanceBudget, revenue, debt, fiscal position2019–2024
World BankHDI, poverty, inequality, world prices2020–2023
IMFCalibration parameters, external sector2020–2024
MMESP v5 — Mongolia Multi-Model Economic Simulation Platform Calibration base year 2024
Core engine · 01

Input–Output

A demand-side Leontief model: shock one sector and watch direct and indirect effects ripple through all 21 sectors of the economy.

Fig. 1 Sectoral impacts — direct vs indirect
Fig. 2 Type I multipliers by sector
Table 1 Top 5 affected sectors
SectorDirect (%)Indirect (%)Total (%)
Fig. 3 Forward vs backward linkages (Rasmussen)
Methodology & equations
$$X = (I - A)^{-1}Y$$

Total output X from technical coefficients A, final demand Y, and the Leontief inverse.

$$m_j = \sum_i l_{ij}$$

Type I multiplier: column sum of the Leontief inverse.

$$R_j = \frac{m_j}{\bar{m}}, \quad D_i = \frac{m_i}{\bar{m}}$$

Rasmussen indices: dispersion of forward and backward linkages.

Model 01 · Leontief Input–Output21 sectors · NSO 2019 I-O table
Core engine · 02

General Equilibrium

A Johansen-linearized CGE where prices, wages and trade adjust together — capturing what partial models miss.

Real GDP impact
—
Equivalent variation
—
Wage change
—
Gov revenue
—
Fig. 1 Sectoral output change
Fig. 2 Sectoral price change
Fig. 3 Trade balance evolution
Methodology & equations
$$\hat{X}_i = a_i + \sum_j \theta_{ij} \hat{X}_j + \sum_k \eta_{ik} \hat{P}_k$$

Johansen hat-algebra: output change as a function of intermediate demand and price changes.

$$\hat{P}_i = \sum_j \alpha_{ij} \hat{P}_j^{in} + \beta_i \hat{w}$$

Price change: weighted input prices plus the wage effect.

Model 02 · Johansen CGE10 sectors · CES production · Armington trade
Core engine · 03

New Keynesian DSGE

A 3-equation quarterly model — IS curve, Phillips curve, Taylor rule — with a commodity shock channel for a mining economy.

Fig. 1 Output gap response
Fig. 2 Inflation response
Fig. 3 Policy rate response
Table 1 Calibration parameters
ParameterValueDescription
Discount factor (β)0.99Quarterly household discount rate
Calvo frequency (θp)0.7525% of firms adjust prices per quarter
Inverse Frisch (φ)1.5Labor supply elasticity
Policy weight (φπ)1.5Central bank inflation reaction
Policy persistence (ρr)0.8Interest rate smoothing
Methodology & equations
$$y_t = E_t[y_{t+1}] - \frac{1}{\sigma}(i_t - E_t[\pi_{t+1}] - r_t^*)$$

IS curve: output gap depends on the real interest rate and the natural rate.

$$\pi_t = \beta E_t[\pi_{t+1}] + \kappa y_t + u_t$$

Phillips curve: inflation driven by the output gap and cost-push shocks.

$$i_t = r_t^* + \pi_t + \phi_\pi(\pi_t - \bar{\pi}) + \phi_y y_t + \epsilon_t^i$$

Taylor rule: the policy rate responds to inflation and the output gap.

Model 03 · New Keynesian DSGEQuarterly · 12-quarter impulse responses
Projections · 04

Macro-Fiscal Framework

A 10-year fiscal path with mining revenue decomposition, debt dynamics and a 60% statutory debt ceiling.

Fig. 1 GDP projection
Fig. 2 Debt-to-GDP vs 60% ceiling
Fig. 3 Revenue vs expenditure
Table 1 Fiscal projections, 10-year
YearGDP (T MNT)RevenueExpenditureBalanceDebt/GDP (%)
Methodology & equations
$$D_{t+1} = (1 + r) D_t + G_t - T_t$$

Debt dynamics: accumulated deficits plus interest cost.

$$\frac{d}{dy} \left(\frac{D}{Y}\right) = \frac{r - g}{1+g} \cdot \frac{D}{Y} + \frac{PB}{Y}$$

Debt-to-GDP is stable when the primary balance offsets the interest–growth differential.

Model 04 · Macro-Fiscal Framework10-year horizon · 60% debt ceiling
Projections · 05

System Dynamics

Stock-flow projections of population, rural–urban migration and urbanization over a 20-year horizon.

Fig. 1 Population projection, 20 years
Fig. 2 Urbanization trend, 20 years
Diagram Migration flows
Rural Region Urban Centers International | | | +----> Rural Pop ----->| | | Flow | | | +----> Urban Pop ------->| | | Emigration | |<---- Return Migration|<-------- Return -------+ | | +----> Rural Births +----> Urban Births | (TFR=2.1) | (TFR=1.8) | | +----> Rural Deaths +----> Urban Deaths (Mortality) (Mortality)
Table 1 Key parameters
ParameterRuralUrban
Total fertility rate (TFR)2.11.8
Mortality rate (/1000)7.25.8
Rural-to-urban migration2.3% / yr
International emigration0.8% / yr
Methodology & equations
$$\frac{dx_i}{dt} = \text{inflow}_i - \text{outflow}_i$$

Each population stock changes by births + immigration minus deaths + emigration.

$$P(t) = P_0 e^{rt}$$

Baseline projection with net growth rate r.

Model 05 · System Dynamics20-year stock-flow projection
Distribution & risk · 06

SAM & Income Distribution

A social accounting matrix that answers the question every aggregate hides: who actually bears the shock, quintile by quintile.

Fig. 1 Income share by quintile
Fig. 2 Lorenz curve — Gini 0.34
Table 1 Distributional impact of a mining shock
Income quintileBaseline share (%)Shock impact (%)New share (%)
Q1 — poorest5.2−3.15.04
Q29.8−2.29.58
Q3 — middle15.4−1.515.17
Q422.6−0.822.42
Q5 — richest47.0−0.346.79
Table 2 Household welfare impacts
Household typeConsumption change (%)Equivalent variation (M MNT)
Agricultural smallholders−4.2−850
Urban low-wage workers−2.8−620
Middle-class professionals−1.1−380
Mining sector workers−6.5−2,150
Government employees−0.4−120
Methodology & equations
$$M = (I - A)^{-1} \cdot B$$

SAM multiplier: Leontief inverse post-multiplied by the income distribution matrix B.

$$\text{Gini} = 1 - 2\sum_{i=1}^{n} (p_i - L_i)$$

Gini from the Lorenz curve: area between the 45° line and the observed curve.

Model 06 · Social Accounting Matrix5 quintiles · 5 household types
Distribution & risk · 07

Debt Sustainability

IMF/World Bank DSA methodology: baseline debt trajectory against four stress scenarios and the 60% statutory ceiling.

Fig. 1 Debt trajectory under stress scenarios
Assessment Risk classification

Baseline scenario  Medium risk

Debt/GDP stable at 44–46% through the forecast period, but vulnerable to commodity shocks given mining at 25.6% of GDP.

Table 1 Stress test comparison
ScenarioPeak debt/GDPYear of peakRisk level
Baseline45.2%2034Medium
Copper −30%58.3%2031High
Growth −2%52.1%2032High
Interest rate +3%51.8%2033High
Combined shock68.5%2030High
Methodology & equations
$$\frac{\Delta D}{D} = \frac{(r - g)D + PB}{Y}$$

Debt growth depends on the interest–growth differential and the primary balance. Sustainability requires PB ≥ −(r−g)·D/Y.

$$\text{DSA Risk} = f(\text{Debt Level}, \text{Commodity Volatility}, \text{Fiscal Space})$$

Risk combines debt metrics, commodity dependence (25.6% of GDP) and fiscal cushion.

Model 07 · Debt Sustainability AnalysisIMF/WB framework · 4 stress scenarios
Development · 08

SDG Tracker

Five development indicators tied back to the macro engine — so a commodity crash shows up not just in GDP, but in poverty and schooling.

Human Development Index
SDG 3 · 4 · 5 — health, education, gender
0.741 — very high
Multidimensional Poverty Index
SDG 1 — no poverty
0.140 — moderate
Gini coefficient
SDG 10 — reduced inequalities
0.34 — moderate inequality
Innovation index
SDG 9 — industry & innovation
38.2 / 100
Poverty rate
SDG 1 — no poverty
18.4% national
Urban air quality index
SDG 13 · 3 — climate, health
52 — moderate
Table 1 Mining −30% shock — SDG impact
IndicatorBaselineShockChange
Poverty rate18.4%24.1%+5.7 pp
Gini coefficient0.3400.371+0.031
HDI0.7410.722−0.019
School enrollment88.3%85.1%−3.2 pp
Health expenditure / capita$68$58−14.7%
Reference SDG coverage
GoalTracked dimensions
SDG 1 — No povertyPoverty rate, inequality, social protection
SDG 3 — Good healthHealth spending, life expectancy, maternal mortality
SDG 4 — Quality educationEnrollment, literacy, education spending
SDG 5 — Gender equalityFemale employment, political representation
SDG 9 — Industry & innovationR&D spending, patents, infrastructure
SDG 10 — Reduced inequalitiesGini, income ratios, social mobility
SDG 13 — Climate actionGHG emissions, renewables, degradation
Model 08 · SDG Indices5 indicators · linked to macro engine
Assistant

Policy Chat

Ask about shocks, sectors and scenarios — answers are grounded in the platform's eight models.

Policy assistantPattern-matched to 8 model engines