## _wp1128

## Source details

**Canonical URL:** [_wp1128](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1128.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1128.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1128.pdf.json)

---

### Model overview
- Small open economy DSGE (New Keynesian) model for Jordan; log-linear approximation and Bayesian estimation (Kalman filter for likelihood; Metropolis-Hastings for posterior).
- Features:
  - Habit formation in utility (parameter h).
  - Calvo-style staggered wage and price setting with indexation parameters (ϑ_u, ϑ_w, Ϧ_u, Ϧ_w).
  - Imperfect competition (Dixit-Stiglitz).
  - Oil explicitly in consumption and production (oil consumption ̂c_oil; oil in production α).
  - Exogenous commodity export sector (potash/phosphate ≈ 4 percent of GDP and 12 percent of total exports).
  - No explicit fiscal sector; no modeled capital accumulation.
- Monetary policy:
  - Taylor-type rule with interest rate inertia (ρ_R), responses to inflation (ω_π), GDP growth (ω_y) and real exchange rate (ω_Δe).
  - Rule allows interest rate smoothing and response to real exchange rate movements.

### Data and estimation setup
- Sample: quarterly Jordanian data 1992:1 to 2009:4.
- Observables used (seven): real GDP; short-term real interest rate; consumer price inflation (CPI); real exchange rate; nominal exchange rate devaluation; real wages; labor input.
- Additional series: oil imports and real price of oil (WTI deflated by external price index).
- Data treatment:
  - Real GDP, consumer prices, real wages, and labor input seasonally adjusted.
  - Headline inflation used as CPI and to deflate nominal wages and construct real exchange rate.
  - Variables demeaned; real wages and GDP detrended and demeaned using a linear trend.
  - Short-term real interest rate = monetary policy rate − expected inflation implicit in CBJ forecast.
  - Labor input = fraction of total employment over working-age population.
- Stochastic drivers: nine orthogonal AR(1) shocks (domestic productivity, foreign interest rate, foreign demand, foreign inflation, labor preference, domestic oil price, international oil price, foreign demand for domestic commodity, monetary shock).

### Parameterization and priors (selected)
- Baseline calibrated values (Table 1):
  - β = 0.99
  - σL = 1.00
  - h = 0.50
  - gy = 3.50
  - δ = 0.10
  - η = 0.20
  - 1-γ = 0.60
  - θ = 1.00
  - Probability of adjusting wages = 0.75
  - Wage indexation/weight of past inflation = 0.50
  - Probability of adjusting (prices) = 0.75
  - Domestic goods indexation at home = 0.50
  - α = 0.40
  - ω = 0.30
  - εL = 9.00
  - Net exports to GDP ratio = 0.27
  - γ* = 1.00
  - Elasticity of FX borrowing (supply) = 0.001
  - Share of potash/phosphate in total exports = 0.12
  - Monetary policy baseline: ρ = 0.75; ωπ = 0.75; ωy = 0.70; ωΔe = 0.70
- Priors (selected):
  - σL: truncated normal mean 1.0, St. Dev. 0.3.
  - h: truncated normal mean 0.5, St. Dev. 0.25.
  - Calvo probabilities ϑ_u and ϑ_w: gamma mean 0.75, St. Dev. 0.05.
  - θ: inverse gamma mean 1.0, St. Dev. 0.3.
  - ω (elasticity of international supply of funds): inverse gamma mean 0.75, St. Dev. 0.2.
  - Policy-rule coefficients ω_π and ω_y: truncated normal mean 0.75, St. Dev. 0.15.
  - ω_Δe: inverse gamma mean 0.5, St. Dev. 0.15.
  - Interest rate smoothing ρ_i: gamma mean 0.75, St. Dev. 0.2.
  - Substitution of oil in consumption and production, η and ω: inverse gamma means 0.15 and 0.10, St. Dev. 0.5.
  - AR(1) persistence priors: gamma mean 0.7, St. Dev. 0.25.

### Bayesian estimation results (posterior means and modes, selected)
- Posterior mode and posterior means reported for fixed exchange rate regime (Table 2, selected entries):
  - σL (Mode) = 0.9529; prior mean 1.0000, St. Dev. 0.3000.
  - h (Mode) = 0.4871; prior mean 0.5000, St. Dev. 0.2500.
  - θ (Mode) = 0.9730; prior mean 1.0000, St. Dev. 0.3000.
  - η (Mode) = 0.1322; prior mean 0.1500, St. Dev. 0.0500.
  - η* (Mode, foreign demand elasticity) = 0.1607; prior mean 0.1000, St. Dev. 0.0100.
  - δ (Mode, share of oil in consumption) = 0.0771; prior mean 0.1000, St. Dev. 0.0100.
  - Risk premium on FX borrowing (Mode) = 0.7350; prior mean 0.7500, St. Dev. 0.2000.
  - ω (producers, Mode) = 0.0865; prior mean 0.1000, St. Dev. 0.5000.
  - α (Mode) = 0.2893; prior mean 0.3000, St. Dev. 0.1000.
  - Price rigidity фH (Mode) = 0.7027; prior mean 0.7500, St. Dev. 0.0500.
  - Wage rigidity фL (Mode) = 0.7436; prior mean 0.7500, St. Dev. 0.0500.
  - ξH (Mode) = 0.1108; prior mean 0.1000, St. Dev. 0.0100.
  - ξL (Mode) = 0.0496; prior mean 0.1000, St. Dev. 0.0100.
  - Monetary policy: ρ (Mode) = 0.6832; ωπ (Mode) = 0.7514; ωy (Mode) = 0.7201; ωΔe (Mode) = 0.4893.
  - Shock persistences (selected modes ≈ prior mean 0.7000): ρah Mode 0.7029; ρs Mode 0.6941; ρc* Mode 0.7070; ρi* Mode 0.7196; ρπ* Mode 0.6692; ρζ Mode 0.6928; ρψ Mode 0.6768.
- Additional posterior point estimates (from text summary):
  - Elasticity of labor supply σL = 0.95.
  - Habit formation h = 0.49.
  - Elasticity between home and foreign goods θ = 0.97.
  - Demand elasticity of home goods abroad η* = 0.16.
  - Estimated risk premium on foreign borrowing = 0.74 under fixed exchange rate.
  - Elasticity of substitution between oil and core consumption η ≈ 0.13.
  - Elasticity between labor and oil in production ω = 0.09.
  - Weight of oil in production α = 0.29.
  - Weight of oil in consumption ϖ (delta) = 0.08.
  - Calvo probabilities (posterior means): prices ϑ_u = 0.70 (re-optimize every 4 quarters); wages ϑ_w = 0.74 (re-optimize every 3.5 quarters).
  - Wage indexation Ϧ_w = 0.05.
  - Price indexation Ϧ_p = 0.11.
  - Interest rate smoothing ρ_i = 0.68.
  - Policy-rule coefficients: ω_y = 0.75; ω_π = 0.72; ω_Δe = 0.49.

### Impulse response and shock effects (baseline, fixed peg)
- Impulse responses correspond to a 1 standard deviation innovation in period 1; price and wage inflation, nominal and real interest rates are annualized growth rates; 90 percent confidence intervals reported.
- Foreign demand shock:
  - Raises income and consumption.
  - Real exchange rate appreciates.
  - Current account/NFA position deteriorates (increase in foreign debt).
  - Imports rise faster than exports.
  - Firms demand more labor; real wages rise given low wage indexation.
  - Domestic prices rise, prompting monetary tightening (nominal interest rate up); export growth dampened.
- Foreign interest rate shock:
  - Contracts consumption and output sharply.
  - Employment and real wages fall; core and headline inflation decline.
  - Imports fall; central bank lowers policy rate gradually.
  - Real exchange rate depreciates; foreign debt falls; current account/NFA improves.
- Monetary-policy shock (one-off increase in interest rate; Jordan: ݄ = 0.49):
  - Generates hump-shaped consumption due to inertia and nominal frictions.
  - Import growth and foreign debt rise; intertemporal positive effect does not dominate negative intratemporal effect.
- International and domestic oil price shocks:
  - Large negative income effect; total consumption and output contract on impact.
  - Demand for consumption goods falls (particularly foreign goods); imports fall substantially.
  - Firms face higher marginal costs, shed labor; real wages and prices fall quickly after an initial inflation uptick.
  - Monetary policy lowers nominal interest rate in response to contraction and initial inflation uptick.
  - Real exchange rate depreciates; current account improves; foreign debt reduced.
  - International shock larger in magnitude than domestic oil price shock; qualitative dynamics similar.
- Positive commodity (potash/phosphate) production shock:
  - Increases commodity output and consumption.
  - Expansion is inflationary; central bank tightens.
  - Real appreciation occurs; net exports growth does not necessarily rise because appreciation harms non-commodity exports.
- Additional notes:
  - Impulse responses under a hypothetical flexible exchange rate are broadly similar but of much lower magnitude; a flexible regime could lower consumption and output volatility relative to external shocks (alternative results available upon request).
  - Hump-shaped real exchange rate response (peak ≈ one year) attributed to inclusion of the risk premium.

### Policy implications and interpretation
- Presence of wage and price rigidities generates a trade-off between stabilizing the output gap and price/wage inflation; strict inflation targeting can be suboptimal when wages are sticky.
- Optimal policy may minimize volatility of a weighted average of wage and price inflation rather than solely targeting price inflation.
- Large estimated response of the interest rate to real exchange rate volatility and significant interest rate smoothing reflect policy under a pegged exchange rate with imperfect asset substitution.
- Pegged exchange rate regime affords a lower risk premium relative to a hypothetical floating regime (estimated risk premium increases by 13 percent under a floating specification as noted in the source).

### Robustness checks (three approaches)
- Robustness exercises performed:
  1. Set all parameter prior distributions to truncated Normal.
     - Example changes: elasticity of foreign demand from 0.16 to 0.20; domestic interest rate smoothing parameter from 0.68 to 0.60. Estimation results do not reveal substantial differences from baseline.
  2. Expand prior standard deviation by 0.05.
     - Posteriors do not differ much from baseline; differences mix in both directions.
  3. Increase prior means by 5 percent.
     - Most posterior estimates shift upward but some (e.g., elasticity of foreign demand) shifted downward.
- Conclusion: model parameters are stable and robust to distributional assumptions on priors.

### Selected exact numeric/statistical values reported
- 0.016 and 0.51, respectively for Latvia.
- 90 percent confidence intervals.
- 1 standard deviation increase of innovation in the initial period.
- Jordan: ݄ = 0.49.
- Elasticity of foreign demand example: 0.16 to 0.20.
- Domestic interest rate smoothing parameter example: 0.68 to 0.60.
- Prior mean shifts in robustness check: increase prior means by 5 percent.
- Expand prior standard deviation by 0.05.
- Nominal wages adjusted optimally every four quarters, on average.
- Prices re-optimized every three and a half quarters, on average.

*Source: _wp1128 - Appendix I: Log-linearized Model (PDF).*

### References .............................................................................................................

### _wp1128 - References

### References
- References ................................................................................................................................29

### Tables
- 1. Baseline Parameterization ....................................................................................................38
- 2. Baseline Specification: Results from Posterior Maximization ............................................39
- 3. Robustness Check: Distribution Density Function ..............................................................40
- 4. Robustness Check: Wider Standard Deviation ....................................................................41
- 5. Robustness Check: Increase Prior Means ............................................................................42

### Figures
- 1. Priors ....................................................................................................................................43
- 2. Impulse Response to a Foreign Demand Shock—̂ܿ
௧
כ
 ...........................................................45
- 3. Impulse Response to a Foreign Interest Rate Shock—ଓ̂
௧
כ
 .....................................................46
- 4. Impulse Response to a Monetary Shock—ොݒ
௧
௠
 .....................................................................47
- 5. Impulse Response to an International Oil Price Shock—ො݋
௧
 ................................................48
- 6. Impulse Response to Domestic Oil Price Shock—߰
෠
௧
 .........................................................49
- 7. Impulse Response to a Foreign Demand Shock to Potash—̂ݏ
௧
 ...........................................50
- 8. Impulse Response to a Labor Preference Shock—ߞ
መ
௅,௧
 ........................................................51

*Source: _wp1128 - References (PDF).*

### Appendix I: Log-linearized Model .......................................................................................

### Appendix I: Log-linearized Model

### I. Introduction and purpose
- Presents a small open economy DSGE (New Keynesian) model estimated for Jordan to evaluate monetary policy rules in presence of nominal and real rigidities.
- Emphasizes that staggered wage and price setting generate a trade-off between stabilizing the output gap, price inflation, and wage inflation (Erceg, Henderson, and Levin (2000); Blanchard and Galí (2005)).
- Uses a log-linear approximation of the model and Bayesian estimation methods (Kalman filter for the likelihood; Metropolis-Hastings algorithm for posterior approximation).

### Model structure and key ingredients
- Features included:
  - Habit formation in consumer utility (parameter h).
  - Wage and price rigidities with Calvo-style staggered setting and indexation parameters (߶_u, ߶_w, ߦ_u, ߦ_w).
  - Imperfect competition (Dixit-Stiglitz aggregation for labor and product varieties).
  - Oil explicitly modeled in consumption and production bundles (oil consumption ܥ̂_oil and oil in production ܱ).
  - Commodity export sector (exogenous endowment) representing potash and phosphate (about 4 percent of GDP and 12 percent of total exports).
  - No explicit fiscal sector, no modeled investment inertia or capital.
- Monetary policy rule:
  - Taylor-type rule with interest rate inertia, responses to inflation and GDP growth deviations, and an augmented response to real exchange rate movements (parameters ߸_π, ߸_y, ߸_q, and persistence ρ_R).
  - Rule allows for interest rate smoothing and a response to real exchange rate volatility (following Lubik and Schorfheide (2007)).

### Households and preferences
- Representative Ricardian households maximize expected lifetime utility with:
  - Consumption composite of core (non-fuel) goods and imported fuel: equations (2)–(5).
  - CES aggregator for core consumption between home and foreign goods (equation (3)).
  - Habit formation parameter h; inverse labor supply elasticity σ_L; AR(1) labor preference shock ζ_l.
  - Money in utility (parameter a) and semi-elasticity of money demand μ.
- Financial assets available:
  - Money ࣧ_t, one-period noncontingent foreign bonds B_t^f, and one-period domestic contingent nominal bonds B_t(s).
  - Premium on foreign borrowing Θ(·) depends on aggregate net foreign asset position (steady-state parameterization Θ and elasticity ω).
- Euler equation and foreign bond FOC given by equations (9) and (10), linking domestic interest rate and foreign interest rate with the premium.

### Labor supply and wage setting
- Households supply differentiated labor (Dixit-Stiglitz aggregation, equations (11)–(13)).
- Wage setting:
  - Calvo-type nominal wage rigidity: each period probability of re-optimizing wages is (1 – ϑ_w); parameter ϑ_w measures wage stickiness.
  - Passive updating rule for non-optimizing households with indexation to past CPI inflation and implicit target inflation (Γ_w,t+i, equation (14)).
  - Wage indexation parameter ߦ_w influences how past inflation affects wages.

### Domestic production, inputs, and pricing
- Firms produce differentiated home-good varieties using labor and imported oil only (production function equation (15)):
  - Oil weight in production α and substitution parameter χ between oil and labor determine sensitivity of output and marginal cost to oil-price shocks.
- Cost minimization yields common marginal cost across firms (expression in text).
- Price setting:
  - Calvo price rigidity for firms: probability of re-optimizing price each period is (1 – ϑ_p); non-optimizing firms follow passive updating rule Γ_p (equation (19)).
  - Price indexation parameter ߦ_p captures degree of price indexation to past inflation.
- Firms maximize discounted profits subject to demand for each variety (equations (17)–(18)).

### Foreign sector and real exchange rate
- Exports: home goods and exportable commodity; foreign demand for home goods given by equation (20) with share parameter ξ^f and elasticity η^f.
- Law of one price for home goods sold abroad: P_x^f = P_x.
- Real exchange rate defined as relative price of foreign consumption basket to domestic consumption basket (equation (21)).
- Domestic real price of oil includes passthrough deviations ψ_t: P_oil^dom = ψ_t * P_oil^world (equation (22)); both P_oil^world and ψ_t follow AR(1).
- Commodity production (potash/potash-like) is exogenously determined by stochastic endowment R_com (equation (23)); increase is a windfall and induces real appreciation.

### Monetary policy specification
- Taylor-type rule (equation (24)) with:
  - Responses ߸_π to inflation deviations, ߸_y to GDP growth deviations (long-run responses), and ߸_q to real exchange rate misalignments/volatility.
  - Interest rate smoothing parameter ρ_R (0 < ρ_R < 1).
  - Monetary shock ε_R,t.
- Interpretation:
  - ߸_y → ∞ implies strict output-gap targeting.
  - ߸_π → ∞ implies strict inflation targeting.
  - ߸_q → ∞ implies exchange-rate targeting.
  - Finite ߸_π and positive ߸_q correspond to a managed float.

### Equilibrium conditions and accounting identities
- Labor market clearing: L_t = ∫ l_t(j) dj = L̄_t (equation (26)).
- Home goods market: Y^h_t = C^h_t + X^h_t (equation (27)).
- Aggregate GDP (implicit deflator P_Y,t) and aggregates follow equations (28)–(31).
- Net foreign asset accounting condition provided (equation (32)).
- Government: no public spending; government budget constraint simplified to transfers and seigniorage identity (equation (25)).

### Stochastic processes
- The economy is driven by nine orthogonal AR(1) stochastic shocks representing log-linear deviations from steady state (text indicates nine AR(1) shocks; variables denoted with hat ˆ).

### Estimation strategy
- Bayesian estimation combining priors and likelihood obtained from Kalman filter on the log-linear model; Metropolis-Hastings algorithm used to approximate the posterior.
- Motivations for Bayesian approach:
  - System-based fit to vector of time series.
  - Estimation based on model likelihood (not matching VAR impulse responses).
  - Incorporation of prior information; helps identification and coping with misspecification.
  - Enables model comparison via marginal likelihood.

### Main results from Bayesian estimation (reported findings)
- Similar to other oil-importer DSGE studies, results include:
  - (i) a low degree of substitution and share of oil in the consumption basket and production function;
  - (ii) a smaller elasticity of labor supply;
  - (iii) a smaller habit formation coefficient in consumption;
  - (iv) a higher elasticity of substitution between home and foreign goods in consumption;
  - (v) smaller estimated Calvo probabilities of optimally resetting prices and wages—implying that prices are reset optimally every 3.5 quarters whereas wages are re-optimized every 4 quarters;
  - (vi) relatively low wage indexation;
  - (vii) a significant degree of interest rate smoothing;
  - (viii) the response of the interest rate to inflation’s deviation from target is similar to output growth’s deviation from potential;
  - (ix) the response of the interest rate to real exchange rate volatility is quite large; and finally,
  - (x) the pegged exchange rate regime affords a lower risk premium (relative to a hypothetical floating exchange rate regime).

### Impulse response results (under current peg)
- Foreign demand shocks:
  - Raise income and consumption;
  - Cause real exchange rate appreciation;
  - Deteriorate the current account/net foreign asset (NFA) position.
- Foreign interest rate shocks:
  - Contract consumption and output;
  - Depreciate the real exchange rate;
  - Improve the current account.
- Monetary-policy shocks (with a high domestic interest rate):
  - Induce households to choose a consumption profile with an increasing growth rate of consumption given consumption inertia;
  - Lead to a rise in foreign debt and deterioration in current account position.
- International and domestic oil price shocks:
  - Result in a large negative income effect;
  - Depreciate the real exchange rate;
  - Improve the current account position.
- Note: responses are broadly similar under a hypothetical flexible exchange rate regime but with much less magnitude; a flexible regime could lower consumption and output volatility relative to external shocks (results available upon request as noted in the source).

### Policy implications and interpretation
- Presence of both wage and price rigidities implies a trade-off between stabilizing inflation and output; full inflation stabilization is suboptimal when wages are sticky (can exacerbate output and wage-inflation volatility).
- Optimal policy may minimize volatility of a weighted average of wage and price inflation rather than solely targeting price inflation (Erceg, Henderson, and Levin (2000); Blanchard and Galí (2005)).
- The estimated large response of the interest rate to real exchange rate volatility and significant interest rate smoothing reflect the policy environment under a pegged exchange rate with imperfect asset substitution.
- Pegged exchange rate regime appears to lower the risk premium relative to a hypothetical floating regime.

*Source: _wp1128 - Appendix I: Log-linearized Model (text provided).*

### Appendix I) : a domestic productivity shock (ොܽ

### Appendix I) : a domestic productivity shock (ොܽ

### Summary of model shocks and structure
- The model includes the following structural shocks (as listed): a domestic productivity shock (ොܽ
  ); a foreign interest rate shock (ଓ̂
  ); a foreign demand shock (̂ܿ
  ); a foreign inflation shock (ොߨ
  ); a labor supply preference shock (ߞ
  መ
  ); a domestic oil price shock (߰
  ෠
  ); an international oil price shock (ො݋
  ); a shock to foreign demand of the domestic commodity (potash/phosphate) (̂ݏ
  ); and a monetary policy shock (ොݒ
  ௠
  ).
- Equations (A1) through (A32) form a linear rational expectation system cast in canonical state-space form (see equation numbering and canonical representation in the source).
- The state vector ࢠ
  contains the model’s variables expressed as log-deviations from steady-state, and the innovation vector ࢿ
  contains white noise structural innovations. Matrices Ω
  are non-linear functions of the structural parameters θ.
- The solution is expressed as ࢠ
  = Ω
  (ߴ) ࢠ
  ـ1 + Ω
  (ߴ) ࢝
  (equation (33) in the source).

### Econometric methodology
- Estimation approach:
  - Bayesian estimation techniques are used for model estimation and evaluation, following Rabanal and Rubio-Ramírez (2005), Lubik and Schorfheide (2006) and Adolfson et al. (2005b).
  - The log-linearized model is combined with a measurement equation to relate observables to model variables (equation (34)).
  - Assuming normally distributed white noise innovations, the Kalman filter is used to compute the conditional likelihood function.
  - Prior distributions with density ܘ(ߴ) are updated by the likelihood L(ߴ|Y_T) to obtain the posterior ܘ(ߴ|Y_T) via Bayes’ theorem (equation (35)).
  - Draws from the posterior distribution are generated via Bayesian simulation techniques (Metropolis-Hastings algorithm) to compute posterior means and standard deviations.
- Measurement equation and observables:
  - Observable vector ࢟
    is assumed to be = [ොݕ
    ̂ݎ,
    ඔߨ,
    ௓,
    Δ݁ ̂
    ෞݎ݁ݎ,
    ̂݌,
    ௌ,
    ෞݎݓ,
    ݈,
    መ
    ඔ݋,
    ൟ] (as given in the source).
  - H selects elements of ࢠ
    corresponding to these observables.

### Data
- Sample and frequency:
  - Quarterly Jordanian data for the period 1992:1 to 2009:4.
- Seven observable variables used in estimation:
  - real GDP,
  - the short-term real interest rate,
  - consumer price inflation (CPI),
  - the real exchange rate,
  - nominal exchange rate devaluation,
  - real wages,
  - labor input.
- Additional series utilized:
  - oil imports and the real price of oil (international price of WTI oil deflated by an index of relevant external prices for the Jordanian economy).
- Data treatment:
  - Real GDP, consumer prices, real wages, and labor input are seasonally adjusted.
  - Headline inflation is used as CPI and to deflate nominal wages and construct the real exchange rate.
  - Variables are demeaned. Real wages and GDP are detrended and demeaned using a linear trend.
  - Short-term real interest rate corresponds to the monetary policy rate minus the expected inflation rate implicit in the CBJ’s forecast.
  - Labor input is constructed as the fraction of total employment over the working-age population.

### Parameterization and calibration
- Parameters estimated within vector ߴ include (exact listing as provided): ߪ൛
  ௅
  ߟ,ߟ,ߠ,݄,
  כ
  ߶,߷,
  உ
  ߶,
  ௅
  ߦ,
  உ
  ߦ,
  ௅
  ߩ,
  ௜
  ߸,
  గ
  ߸,
  ௬
  ߸,
  ୼௘
  ߩ,
  ௔
  ߩ,
  ఍
  ߩ,
  ௖
  כ
  ߩ,
  ట
  ߩ,
  ௜
  כ
  ߩ,
  గ
  כ
  ߩ,
  ௦
  ߪ,
  ௔
  ߪ,
  ఍
  ߪ,
  ௖
  כ
  ߪ,
  ట
  ߪ,
  ௜
  כ
  ߪ,
  గ
  כ
  ߪ,
  ௦
  ߪ,
  ఔ
  ൟ.
- Parameters estimated outside the model using WEO oil data:
  - ߩ
    ௢ (persistence of real international price of oil) = 0.92.
  - ߪ
    ௢ (variance of the oil price shock) = 12.4 percent.
- Assumptions and calibrations:
  - ߩ
    ఔ (persistence of monetary shock) is assumed to be zero.
  - Annual long-run labor productivity growth assumed to be 3.5 percent (consistent with 6 percent long-run GDP growth and 2 percent labor force growth).
  - Long-run annual inflation rate is 6 percent.
  - Subjective discount factor β is set close to 0.99 (quarterly basis) to yield an annual nominal interest rate of 7.5 percent in the steady state.
  - Share of imported goods in the consumption basket, ߛ, = 40 percent.
  - Share of home goods production in total GDP, (ܥ
    உ
    ܥ
    ൅
    உ
    ) ܻ⁄ = 90 percent.
  - Ratio of net imports to GDP, (ܺെܯ) ܻ⁄ = 25 percent in the steady state.
  - Elasticity of substitution between different types of labor, ߳
    உ
    உ
    = 11 (value used within the range from other studies).
  - Natural commodity resources, S, account for the remaining 10 percent of GDP (footnote 15).
- Oil shares and elasticities:
  - Total oil imports ratio to GDP, (ܱ
    ௖
    ܱ
    ൅
    உ
    ) ܻ⁄ ≈ 0.15; steady-state share of oil in production of home goods computed by subtracting household fuel consumption.
  - Estimated autoregressive process parameters for international oil price: ߩ
    ௢ = 0.92 and ߪ
    ௢ = 12.4 percent (reiterated).

### Prior distributions
- Priors are informed by evidence from other oil-importer DSGE studies (Chile, Hungary, Latvia, Mozambique) and diffuse where evidence is weak.
- Specific priors highlighted:
  - Inverse elasticity of labor supply, ߪ
    ௅: truncated normal distribution with mean 1.0 and standard deviation 0.3.
  - Habit formation coefficient, h: truncated normal distribution with mean 0.5 and standard deviation 0.25.
  - Calvo probabilities ߶
    உ and ߶
    ௅: gamma distribution with mean 0.75 and standard deviation 0.05.
  - Elasticity of substitution between foreign and domestic goods, θ: inverse gamma distribution with mean 1.0 and standard deviation 0.3.
  - ߟ
    כ: truncated normal distribution with mean 0.1 and standard deviation 0.01.
  - Elasticity of the international supply of funds, ߷: inverse gamma distribution with mean 0.75 and standard deviation 0.2.
  - Policy-rule coefficients ߸
    గ and ߸
    ௬: truncated normal distributions with mean 0.75 and standard deviation 0.15.
  - ߸
    Δ௘: inverse gamma distribution with mean 0.5 and standard deviation 0.15.
  - Interest rate smoothing coefficient, ߩ
    ௜: gamma distribution with mean 0.75 and standard deviation 0.2.
  - Substitution of oil in consumption and production, η and ω: inverse gamma distributions with mean 0.15 and 0.10, respectively, and the same standard deviation 0.5.
  - Autoregressive parameters (persistence) ߩ
    ఍, ߩ
    ௜
    כ, ߩ
    గ
    כ, ߩ
    ௌ, ߩ
    ௖
    כ, ߩ
    ట, ߩ
    ௔: gamma distributions with prior mean 0.7 and standard deviation 0.25 (diffuse priors).

### Bayesian estimation results (posterior means and interpretation)
- Estimation procedure: compute posterior mode and then posterior distribution with Metropolis-Hastings algorithm. Posterior means under the fixed exchange rate regime are reported (Table 2 referenced).
- Key posterior estimates and interpretations:
  - Elasticity of labor supply, ߪ
    ௅ିଵ = 0.95 (implies stronger labor supply sensitivity to real wages than U.S., Chile, Mozambique estimates).
  - Habit formation coefficient, h = 0.49 (implies autoregressive coefficient for consumption h/(1 - h) ≈ 0.95).
  - Elasticity of substitution between home and foreign goods in consumption basket, θ = 0.97.
  - Demand elasticity of home goods abroad, ߟ
    כ = 0.16 (implies limited price elasticity of foreign demand).
  - Estimated risk premium facing Jordan on foreign borrowing = 0.74 under fixed exchange rate regime.
  - Elasticity of substitution between oil and core consumption, η ≈ 0.13.
  - Elasticity between labor and oil in production, ω = 0.09.
  - Weight of oil in production, ߙ = 0.29; weight of oil in consumption, ߜ = 0.08.
  - Calvo probabilities (posterior means):
    - Home goods prices, ߶
      உ = 0.70 (implying average re-optimization every 4 quarters).
    - Domestic wages, ߶
      ௅ = 0.74 (implying average re-optimization every 3.5 quarters).
  - Wage indexation coefficient, ߦ
    ௅ = 0.05 (implying relatively low wage indexation).
  - Price indexation, ߦ
    உ = 0.11 (no significant evidence of large price indexation).
  - Interest rate smoothing coefficient, ߩ
    ௜ = 0.68.
  - Policy-rule coefficients:
    - Response to GDP deviation, ߸
      గ = 0.75.
    - Response to inflation deviation, ߸
      ௬ = 0.72.
    - Response to real exchange rate volatility, ߸
      Δ௘ = 0.49.
- Comparative notes:
  - The estimated persistence and rigidities are compared to other studies (Chile, euro area, Hungary, U.S.), with Jordan showing lower elasticities of substitution involving oil and lower price/wage indexation relative to some advanced economies.
  - The estimated risk premium increases by 13 percent under a hypothetical floating exchange rate specification (alternative results noted as available upon request).
- Data limitations:
  - Missing data for real wages and employment in Jordan were randomly generated; results should be taken with caution (explicitly noted in source).

*Source: Appendix I of the provided IMF working paper content unit.*

### 0.016 and 0.51, respectively for Latvia.

### _wp1128 - 0.016 and 0.51, respectively for Latvia.

### V. EFFECTS OF SHOCKS — Baseline Results
- Method:
  - Impulse responses estimated with posterior mean parameters under the current fixed exchange rate cases; results reported in Figures 2–8 and Table 2.
  - Posterior distributions of impulse responses constructed by drawing parameters and shock variances from posterior distributions; 90 percent confidence intervals reported.
  - Impulse responses correspond to a 1 standard deviation increase of innovation in the initial period. Price and wage inflation, nominal and real interest rates are defined as annualized growth rates.
- Shocks analyzed (posterior impulse responses reported):
  - foreign demand shock (̂ܿ
௧
כ
)
  - foreign interest rate shock (ଓ̂
௧
כ
)
  - monetary policy shock (ොݒ
௧
௠
)
  - international oil price shock (ො݋
௧
)
  - domestic oil price shock (߰
෠
௧
)
  - shock to foreign demand of the domestic commodity (potash/phosphate) (̂ݏ
௧
ሻ
  - labor supply preference shock (ߞ
መ
௅,௧
)
- Key baseline simulation results:
  - Foreign demand shock:
    - Raises domestic production of home goods and generates a larger positive income effect/expansion in total output on impact.
    - Raises total consumption, tilted more toward foreign consumption goods.
    - Consumption of imported oil increases more toward consumption than production.
    - Imports rise faster than exports; current account/NFA position deteriorates (increase in foreign debt).
    - Firms demand more labor; with relatively low indexation of wages, real wages rise.
    - Domestic prices—core and headline inflation and real home goods prices—rise, prompting monetary tightening (nominal interest rate up) and inducing real exchange rate appreciation, dampening export growth.
  - Foreign interest rate shock:
    - Sharply contracts consumption and output on impact.
    - Employment/labor falls; downward pressure on real wages and core and headline inflation.
    - Imported oil and goods fall; central bank gradually lowers its policy rate.
    - Firms shift production toward exportation; real exchange rate depreciates; foreign debt falls; current account/NFA improves.
  - Monetary-policy shock (one-off increase in the interest rate; Jordan: ݄ = 0.49):
    - Generates a hump-shaped consumption profile due to inertial behavior and nominal frictions.
    - Import growth and foreign debt rise; intertemporal positive effect does not dominate negative intratemporal effect in Jordan.
  - International and domestic oil price shocks:
    - Negative income effect contracts total consumption and output on impact.
    - Demand for all consumption goods falls, particularly foreign goods; imported oil and home goods fall less due to low substitution elasticities.
    - Firms face higher marginal costs from higher real oil price, shed labor; real wages and prices fall fast.
    - Core and headline inflation rise initially then adjust downward as real wages and real home good prices fall.
    - Monetary policy lowers nominal interest rate in response to contraction in output and initial inflation uptick.
    - Firms shift production towards exports aided by lower prices and a depreciating real exchange rate; large fall in imports and some pickup in exports improve current account and reduce foreign debt.
    - No qualitative difference between international and domestic oil price shock other than larger magnitude for the international shock.
  - Positive production shock to exportable commodity (potash/phosphate):
    - Increases production of the commodity, output, and consumption.
    - Expansion is inflationary; higher real domestic prices and wages prompt central bank tightening.
    - Real appreciation occurs; while potash exports rise, net exports growth does not rise because appreciation harms non-commodity exports.

- Additional model behavior and notes:
  - Posterior distributions of impulse responses similar to simulations with posterior mean parameters; confidence intervals are quite narrow (statistically significant).
  - Hump shape of real exchange rate impulse response (peak after about one year) attributed to inclusion of the risk premium.
  - No reported impact of a productivity shock or foreign inflation shock in the text (productivity shock available upon request; foreign inflation shock immaterial given the pegged exchange rate).

### V. EFFECTS OF SHOCKS — Robustness
- Approaches adopted:
  - Comparison of priors and posteriors within the DSGE model (second common robustness approach in the literature).
  - Three robustness checks performed:
    1. Set all parameter prior distributions to a truncated Normal.
       - Result: Estimation results (Table 3) do not reveal substantial differences from baseline.
       - Example parameter changes: elasticity of foreign demand goes from 0.16 to 0.20; domestic interest rate smoothing parameter goes from 0.68 to 0.60.
    2. Expand the standard deviation of the prior distribution by 0.05.
       - Result: Estimation results (Table 4) show posteriors do not differ much from baseline; differences go in both directions.
    3. Increase prior means by 5 percent from benchmark values.
       - Result: Estimation results (Table 5) show most posterior estimates shifted upward but some (e.g., elasticity of foreign demand) shifted downward.
- Conclusion on robustness:
  - Model parameters are stable and robust to distributional assumptions on priors.

### VI. CONCLUSIONS — Main findings and model properties
- Model description:
  - An estimated DSGE model for the Jordanian economy in the New Keynesian tradition; firms adjust prices infrequently and wages are set in a staggered fashion.
  - Oil is an input to production and part of household consumption basket.
  - Flexible elasticity of substitution between oil and other consumption goods in the consumption bundle and in firms' technology.
  - Key structural parameters jointly estimated using a Bayesian approach; estimates fall within plausible ranges.
  - Simulations conducted to evaluate different exchange rate policies.
- Main empirical results:
  - Foreign demand shocks:
    - Raise income and consumption, cause real exchange rate appreciation, and deteriorate the current account/NFA position.
  - Foreign interest rate shocks:
    - Contract consumption and output, depreciate the real exchange rate, and improve the current account.
  - Monetary-policy shocks (high domestic interest rate):
    - Induce households to choose a consumption profile characterized by an increasing growth rate of consumption; foreign debt rises and the current account deteriorates.
  - International and domestic oil price shocks:
    - Produce a large negative income effect, depreciate the real exchange rate, and improve the current account.
  - Price and wage rigidities:
    - Models with both price and wage rigidities best account for the Jordanian data.
    - Degree of wage rigidities is lower than that of domestic prices:
      - Nominal wages are adjusted optimally every four quarters, on average.
      - Prices are re-optimized every three and a half quarters, on average.
    - Low wage indexation generates a less persistent response of inflation to shocks and is not a main determinant of the policy trade-off.
  - Real rigidities:
    - Habit formation and similar real rigidities provide a better account of the aggregate data; estimated values are quantitatively larger than for other net-oil importer models.

### Selected exact numeric/statistical values reported
- 0.016 and 0.51, respectively for Latvia.
- 90 percent confidence intervals.
- 1 standard deviation increase of innovation in the initial period.
- Jordan: ݄ = 0.49.
- Elasticity of foreign demand example: 0.16 to 0.20.
- Domestic interest rate smoothing parameter example: 0.68 to 0.60.
- Prior mean shifts in robustness check: increase prior means by 5 percent.
- Expand prior standard deviation by 0.05.
- Nominal wages adjusted optimally every four quarters, on average.
- Prices re-optimized every three and a half quarters, on average.

*Source: _wp1128 - 0.016 and 0.51, respectively for Latvia.*

### APPENDIX I

### APPENDIX I

### Log-linearized Model — setup and notation
- Model is log-linearized using Taylor expansions around the steady state.
- Steady state normalizations:
  - Productivity normalized to ܣ
ு
ൌ
ఢ
ହ
ఢ
ହ
ିଵ
.
  - Steady state labor disutility parameter ߞ normalized so that the real wage is one.
  - Under these normalizations and proper choice of foreign currency price level of imported goods all relative prices are one.
- Notation:
  - Lowercase with hat (^) denotes log deviation from steady state.
  - Real price of good J: ෞݎ݌
௃,௧
; corresponding nominal price relative to consumption bundle: ෞݎ݌
௃,௧
̂݌ൌ
௃,௧
̂݌െ
௧
.
  - Real wage corresponds to nominal wage relative to CPI: ෞݎݓ
௧
ෝݓൌ
௧
̂݌െ
௧
.
  - For a pegged exchange rate regime set nominal exchange rate movement to naught: ∆݁̂ ൌ0.

### A.1 Aggregate Demand
- Detrended and log-linearized domestic consumption of home and foreign goods, and oil consumption:
  - ̂ܿ
ு,௧
ൌ
ሺ
ߛ1െ
ሻሺ
ߟെߠ
ሻ
ෞݎ݁ݎ
௧
െ
ሺ
ߠ
ሺ
ߛ1െ
ሻ
ߟߛ൅
ሻ
ෞݎ݌
ு,௧
̂ܿ൅
௧
  (A1)
  - ̂ܿ
ி,௧
ߟ൅ߛߠൌେ൫
ሺ
ߛ1െ
ሻ
ෞݎ݁ݎ൯
௧
̂ܿ൅
௧
  (A2)
  - ෙො݋
஼,௧
ෞ  ݎ݌ߟൌେ
ை,௧
̂ܿ൅
௧
  (A3)
- Real exchange rate log-deviation: ෞݎ݁ݎ
௧
̂݁ൌ
௧
̂݌൅
ி,௧
כ
̂݌െ
௧
.
- Law of one price assumed for imported good: ̂݌
ி,௧
ൌ ݁̂
௧
̂݌൅
ி,௧
כ
, where ̂݌
ி,௧
כ
is imported good price in foreign currency.
- Euler equation and uncovered interest parity (log-linear):
  - ̂ܿ
௧
ൌ 
ଵ
ଵା௛
ܧ
௧
ሺ
̂ܿ
௧ାଵ
ሻ
൅
௛
ଵା௛
̂ܿ
௧ିଵ
େ
ଵି௛
ଵା௛
ሾ
ଓ̂
௧
ܧେ
௧
ሺ
ොߨ
௧ାଵ
ሻሿ
  (A4)
  - ଓ̂
௧
ൌ ଓ ̂
௧
כ
ܧ൅
௧
ሺ
̂݁∆
௧ାଵ
ሻ
ܾ߷൅
෠
௧
כ
  (A5)
- Foreign interest rate process (including risk premium):
  - ଓ̂
௧
כ
ߩൌ
௜
כ
ଓ̂
௧ିଵ
כ
ߝ൅
௜
כ
, (A6)

### A.2 Aggregate Supply and Inflation
- Inflation of home goods from optimal price setting and passive resetting price equation (19):
  - ෙොߨ
ு,௧
ൌ
ሺ
߶1െ
ு
ሻሺ
߶ߚ1ெ
ு
ሻ
߶
ு
ሺ
ߦߚ1൅
ு
ሻ
ቀ
ሺ
ߙ1െ
ሻ
ෞݎݓ
௧
ෞ ݎ݌ߙ൅
ை,௧
ොߙெ
ு,௧
ෞ ݎ݌ெ
ு,௧
ቁ 
൅
ఉ
ଵାఉక
హ
ܧ
௧
ොߨ
ு,௧ାଵ
൅
క
హ
ଵାఉక
హ
ොߨ
ு,௧ిଵ
  (A7)
- Firms' cost-minimization relation between inputs and relative prices:
  - ෙො݋
ு,௧
݈ெ
መ
௧
ෞ  ݎݓ൫߱ൌ
௧
ෞ ݎ݌ெ
ை,௧
൯  (A8)
- Log-linearized output in home goods sector from production function:
  - ෙොݕ
ு,௧
ොܽൌ
ு,௧
൅
ሺ
ߙ1ெ
ሻ
݈
መ
௧
ො ݋ߙ൅
ு,௧
  (A9)
- Technology in home goods sectors (AR(1)):
  - ෙොߙ
ு,௧
ߩൌ
௔
ොߙ
ு,௧ିଵ
ߝ൅
௔,௧
  (A10)
- Real wages (log-linear) combining wage-setting and updating rule:
  - (See expression labeled A11 in source; includes parameters ߭, ߞ, and preference shock ߞ
መ
௧ which follows:)
  - ߞ
መ
௧
ߩൌ
఍
ߞ
መ
௧ିଵ
ߝ൅
఍,௧  with ܧ
௧ିଵ
ߝ൫
఍,௧
൯ൌ0 and ܧ
௧ିଵ
ߝ൫
఍,௧
ଶ
ߪ൯ൌ
఍
ଶ
  (A12)
- Marginal rate of substitution between labor and consumption:
  - ෞݏݎ݉
௧
ߪൌ
௅
݈
መ
௧
൅
ଵ
ଵି௛
̂ܿ
௧
െ
௛
ଵି௛
̂ܿ
௧ିଵ
  (A13)

### A.3 Relative Prices
- Real price of home goods and domestic currency real price of oil:
  - ෞݎ݌ 
ு,௧
ෞ ݎ݌ൌ
ு,௧ିଵ
ොߨ൅
ு,௧
ොߨෙ
௧
, (A14)
  - ෞݎ݌
ை,௧
ෞ ݎ݁ݎൌ
௧
ෞ ݎ݌ପ
ை,௧
כ
߰൅
෠
௧
. (A15)
- Real price of oil abroad (AR(1)):
  - ෞݎ݌
ை,௧
כ
ߩൌ
௢
ෞݎ݌
ை,௧ିଵ
כ
ߝ൅
௢,௧
  (A16)
  - ܧ
௧ିଵ
ߝ൫
௢,௧
൯ൌ0 and ܧ
௧ିଵ
ߝ൫
௢,௧
ଶ
ߪ൯ൌ
௢
ଶ
- Deviation of law of one price for oil follows AR(1):
  - ߰
෠
௧
ߩൌ
ట
߰
෠
௧ିଵ
ߝ൅
ట,௧  (A17)
- Foreign inflation in foreign currency and real exchange rate relation:
  - ෞݎ݁ݎ
௧
ෞ ݎ݁ݎൌ
௧ିଵ
൅Δ݁ ̂
௧
ොߨ൅
௧
כ
ොߨෙ
௧
  (A18)
- Foreign inflation exogenous process (AR(1)):
  - ෙොߨ
௧
כ
ߩൌ
గ
ොߨ
௧ିଵ
כ
ߝ൅
గ
כ
, (A19)
  - ܧ
௧ିଵ
ߝ൫
గ
כ
,௧
൯ൌ0 and ܧ
௧ିଵ
ߝ൫
గ
כ
,௧
ଶ
ߪ൯ൌ
గ
כ
ଶ
- CPI and core consumption price level relation with real price of oil, home goods, and real exchange rate:
  - ෞ  ݎ݌ߜ0ൌ
ை,௧
כ
൅
ሺ
ߜ1ெ
ሻ
ෞݎ݌ߛ
ு,௧
൅
ሺ
ߜ1ெ
ሻሺ
ߛ1ெ
ሻ
ෞݎ݁ݎ
௧
  (A20)

### A.4 Aggregate Equilibrium
- Open economy IS curve: market clearing for home goods sector (from A20 and equation (10)):
  - ෙොݕ
ு,௧
ൌቀ
஼
ಹ
௒
̂ܿቁ
ு,௧
൅ቀ
௒
ಹ
ି஼
ಹ
௒
ಹ
̂ܿቁ
௧
כ
ߟെ
כ
ቀ
௒
ಹ
ି஼
ಹ
௒
ஹ
ෞ ݎ݌ቁ൫
ு,௧
ෞ ݎ݁ݎெ
௧
൯  (A21)
  - ஼
ஹ
௒
 corresponds to steady state fraction of home goods consumed by domestic households.
- Total output (log-linearized):
  - ෞොݕ
௧
ൌ
஼
௒
̂ܿ
௧
൅
௑
௒
ොݔ
௧
ෙ
ெ
௒
ෝ݉
௧
  (A22)
  - ஼
௒
: consumption ratio to GDP in steady state.
  - ௑
௒
: total exports to GDP ratio.
  - எ
௒
: total imports to GDP ratio.
- Exports (detrended, log-linearized):
  - ෞොݔ
௧
ߟൌେ
כ
ቀ
஼
ஹ
כ
௑
ෞ ݎ݌ቁ൫
ு,௧
ෞ ݎ݁ݎெ
௧
൯
ෙ
ቀ
஼
ஹ
כ
௑
̂ܿቁ
௧
כ
൅
௒
ೄ
௑
ොݕ
ௌ,௧
  (A23)
- Commodity exports and foreign consumption exogenous AR(1) processes:
  - ෞොݕ
ௌ,௧
ߩൌ
ௌ
ොݕ
ௌ,௧ିଵ
ߝ൅
ௌ,௧  (A24)
  - ̂ܿ
௧
כ
ߩൌ
஼
כ
̂ܿ
௧ିଵ
כ
ߝ൅
௖
כ
,௧  (A25)
- Exports deflator relative to CPI: real price index of exports ෞݎ݌
௫,௧
ൌ
ሺ
ܿ
ு
כ
ܺ
⁄ሻ
ෞݎ݌
ு,௧
, assuming real price of commodity exports constant implies ෞݎ݌
ௌ,௧
ൌ0.
- Imports (detrended, log-linearized) and real price:
  - ෞో݉
௧
ൌ
஼
ಷ
ெ
̂ܿ
ி,௧
൅
ெி஼
ಷ
ெ
ො݋
௧
  (A26)
  - Total oil imports:
    - ෞො݋
௧
ൌ
ை
಴
ை
ො݋
஼,௧
൅
ை
ஹ
ை
ො݋
ு,௧  (A27)
  - Real price index of imports: ෞݎ݌
ெ,௧
ൌ
ܥ
ி
ܯ
ෞݎ݁ݎ
௧
൅
ܱ
ܯ
ෞݎ݌
ை,௧
כ
- Net foreign asset position evolution (long expression labeled A28 in source; includes weights and terms for exports, inflation differentials, and real exchange deviations).

### A.5 Policy Rule
- Linearized baseline policy rule:
  - ̂ݎ
௧
ߩൌ
௜
̂ݎ
௧ିଵ
൅
ሺ
ߩ1ೆ
௜
ሻ
߸൫
గ
ොߨ
௧
߸൅
௬
ොݕ
௧
߸൅
୼௘
̂݁Δ
௧
൯
൅ߥ ̂
௧
  (A29)
  - ̂ݎ
௧
: deviation of the real interest rate from steady state defined as:
    - ̂ݎ
௧
ൌଓ̂
௧
ܧെ
௧
ොߨ
௧ାଵ
  (A30)
- Policy parameters:
  - ߸
π
 and ߸
y
: long-run responses to deviations of inflation and GDP growth from steady state.
  - ߸
Δe
: reaction to real devaluation.
  - ߩ
i
: degree of interest rate smoothing.
- Monetary shock process:
  - ̂ߥ
௧
ߩൌ
ఔ
̂ߥ
௧ିଵ
ߝ൅
ఔ,௧  (A32)

### Parameterization — Baseline Parameter Values (Table 1)
- Households and labour
  - β 0.99 Subjective discount rate (quarterly)
  - σL 1.00 Inverse of the elasticity of labor supply
  - h 0.50 Coefficient of habit formation
  - gy 3.50 Annual productivity growth rate
- Private consumption basket
  - δ 0.10 Share of imported oil in consumption
  - η 0.20 Elasticity of substitution in consumption between core consumption and imported oil
  - 1-γ 0.60 Home bias in core consumption
  - θ 1.00 Intratemporal elasticity of substitution between domestic and foreign goods
- Nominal rigidities
  - 0.75 Probability of adjusting wages
  - 0.50 Wage indexation/weight of past inflation
  - 0.75 Probability of adjusting
  - 0.50 Domestic goods indexation at home
- Domestic production technology
  - α 0.40 Share of imported oil in domestic production
  - ω 0.30 Elasticity of substitution between oil and other factors of production
  - εL 9.00 Elasticity of substitution of different labour varieties
- Foreign sector
  - 0.27 Net exports to GDP ratio
  - γ* 1.00 Price elasticity of foreign demand for domestic goods
  - 0.001 Elasticity of FX borrowing (supply)
  - 0.12 Share of potash/phosphate in total exports
- Monetary policy (baseline)
  - ρ (interest rate smoothing) 0.75
  - ωπ Reaction to inflation 0.75
  - ωy Reaction to output gap 0.70
  - ωΔe Reaction to real exchange rate misalignment 0.70

### Posterior Maximization and Robustness Checks — Selected posterior summaries (Tables 2–5)
- Table 2. Baseline specification: Results from Posterior Maximization (selected entries)
  - Consumers: Habit formation (mean 0.5000, St. Dev. 0.2500, Mode 0.4871)
  - Ѳ: Intratemporal elasticity (Inverse gamma) mean 1.0000, St. Dev. 0.3000, Mode 0.9730
  - η: Elasticity between oil and core consumption (Inverse gamma) mean 0.1500, St. Dev. 0.0500, Mode 0.1322
  - η*: Elasticity of foreign demand (Normal) mean 0.1000, St. Dev. 0.0100, Mode 0.1607
  - δ: Share of oil in consumption (Gamma) mean 0.1000, St. Dev. 0.0100, Mode 0.0771
  - σL: Inverse elasticity of labor supply (Normal) mean 1.0000, St. Dev. 0.3000, Mode 0.9529
  - Risk premium on FX borrowing (Inverse gamma) mean 0.7500, St. Dev. 0.2000, Mode 0.7350
  - Producers ω (Inverse gamma) mean 0.1000, St. Dev. 0.5000, Mode 0.0865
  - α (Gamma) mean 0.3000, St. Dev. 0.1000, Mode 0.2893
  - Rigidity фH (Gamma) mean 0.7500, St. Dev. 0.0500, Mode 0.7027
  - фL (Gamma) mean 0.7500, St. Dev. 0.0500, Mode 0.7436
  - ξH (Gamma) mean 0.1000, St. Dev. 0.0100, Mode 0.1108
  - ξL (Gamma) mean 0.1000, St. Dev. 0.0100, Mode 0.0496
  - Monetary policy: ρ mean 0.7500, St. Dev. 0.2000, Mode 0.6832; ωπ mean 0.7500, St. Dev. 0.1500, Mode 0.7514; ωy mean 0.7500, St. Dev. 0.1500, Mode 0.7201; ωΔe (Inverse gamma) mean 0.5000, St. Dev. 0.1500, Mode 0.4893
  - Shock persistences (selected): ρah mean 0.7000, St. Dev. 0.2500, Mode 0.7029; ρs mean 0.7000, St. Dev. 0.2500, Mode 0.6941; ρc* mean 0.7000, St. Dev. 0.2500, Mode 0.7070; ρi* mean 0.7000, St. Dev. 0.2500, Mode 0.7196; ρπ* mean 0.7000, St. Dev. 0.2500, Mode 0.6692; ρζ mean 0.7000, St. Dev. 0.2500, Mode 0.6928; ρψ mean 0.7000, St. Dev. 0.2500, Mode 0.6768.
- Tables 3–5 report robustness checks with alternative prior densities and wider standard deviations; selected posterior means and modes adjust but retain comparable magnitudes (see source tables for full entries).

### Figures and Impulse Responses (listed)
- Figure 2. Impulse Response to a Foreign Demand Shock—̂ܿ
௧
כ
- Figure 3. Impulse Response to a Foreign Interest Rate Shock—ଓ̂
௧
כ
- Figure 4. Impulse Response to a Monetary Shock—ොݒ
௧
௠
- Figure 5. Impulse Response to an International Oil Price Shock—ො݋
௧
- Figure 6. Impulse Response to a Domestic Oil Price Shock—߰
෠
௧
- Figure 7. Impulse Response to a Foreign Demand Shock to Potash—̂ݏ
௧
- Figure 8. Impulse Response to a Labor Preference Shock—ߞ
መ
௅,௧

*Source: _wp1128 - APPENDIX I*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp1128.pdf_