## _wp0504

## Source details

**Canonical URL:** [_wp0504](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp0504.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp0504.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp0504.pdf.json)

---

### I. INTRODUCTION — framing and motivation
- Government spending in macro models is typically introduced as non-wage government consumption of goods and services.
- Spending for public employment is an important government spending component, while non-wage public consumption tends to show the least significant dynamics.
- Paradox identified: government non-wage consumption often receives the most attention despite limited dynamics (Alesina and Perotti, 1995, p. 224).
- Objective: analyze effects of government spending composition in a two-country general equilibrium sticky-price model to resolve this paradox.
- Literature position and model choice:
  - Exceptions distinguishing public employment vs. non-wage consumption: Finn (1998) and Lane and Perotti (2003).
  - This paper uses a two-country model in the “new open economy macroeconomics” strand rather than a closed-economy flexible-price RBC model (allows study of both long-run full-employment equilibrium and short-run demand-determined output with nominal rigidities; enables analysis of exchange rate and current account effects).

### Model setup and key mechanisms
- Environment:
  - Two-country world (Home and Foreign) with continuum of infinitely-lived agents; total world population normalized to 1. Home agents on interval [0 n], foreign on [1]n,. A measure 1 of firms; n firms domestic, 1-n firms foreign.
  - Agents derive utility from consumption of a composite good, leisure and real balances.
  - Key parameters and variables preserved as in source: β, σ, χ, k, B, tr, tM, C, P, W, tΠ, tτ.
  - Private consumption index and price index specified by equations (4) and (5) with monopolistic competition parameter 1θ>.
  - Law of one price / PPP: P = E P∗.
- Production and government:
  - Labor-only production function: Y = P t L (equation (9)).
  - Relationship between product price and nominal wage: equation (10).
  - Government budget constraint (real terms): equation (11) — government finances non-wage consumption and government employment with lump-sum taxes and seignorage.
- Short-run nominal rigidities and log-linearization:
  - One period preset prices in producer currency (Obstfeld and Rogoff timing); domestic-currency prices of domestic goods and foreign-currency prices of foreign goods rigid in the period of the shock, adjust thereafter.
  - Notation: tildes for short-run log-linearized variables, hats for long-run.
  - Log-linearized system summarized in Table 1 (equations (13)–(24)): price index relation (13), Euler (14), short- and long-run money demand (15),(16), labor-leisure and labor market equilibrium (17),(18), demand and current-account relations (19)–(21), government budget (22), PPP (23), world goods market (24).
  - Initial steady state: government spending for goods and net foreign assets zero for both countries; denoted with zero subscript (0GG BB====).
- Two key equilibrium relationships (graphical solution):
  - Money-demand relationship (short-run relative consumption and exchange rate): equation (25)
    - 1 ()ecc σ =− −   (equation (25) as in source)
    - Interpretation: increase in domestic consumption relative to foreign raises money demand; requires domestic appreciation (decrease in e) to restore money-market equilibrium (“money demand” effect).
  - Current-account relationship (consumption, exchange rate, and policy variables): equation (30)
    - ii * ** 00 000 1111 (1) ()()() 1()1 G gg PP LL K eccllgg LLL θ θδ δθθθ − =+−+−+ − −+−  ll * 1 () (1) gg δθ +− − . (equation (30) as in source)
    - Depends on policy variables (government employment and non-wage consumption); expresses a positive relationship between e and *cc− (worsened current account → depreciation).
  - Graphical apparatus: downward-sloping (25) and upward-sloping (30) curves determine short-run equilibrium in (e, *cc−) space (Figure 1).

### Policy experiments analyzed
- Policy A — Permanent reduction in domestic government employment used to finance a reduction in general taxation:
  - Immediate short-run results:
    - Increase in domestic private consumption compared to foreign private consumption.
    - Domestic appreciation (lower e) in short run.
  - Mechanisms:
    - Agents anticipate lower lifetime tax burden → higher lifetime private wealth → immediate increase in consumption (rational expectations).
    - Sectoral reallocation: public-sector employment reduction compensated by increased private-sector employment in new long-run equilibrium → more productive use of domestic resources.
    - Sectoral shift alleviates current account pressures and contributes to an appreciated short-run exchange rate via expenditure-switching.
- Policy B — Reduction in public employment used to finance an increase in government non-wage consumption:
  - Macroeconomic effects are ambiguous.
  - Domestic agents do not gain from a lower tax burden, but benefit from expanded demand due to higher public consumption.
  - Stronger pressure for reduction in domestic relative private consumption emerges when the initial share of private income derived from employment in the public sector is larger.

### Short-run effects of public employment retrenchment (Section III.B)
- Permanent asymmetric decrease in domestic government employment relative to foreign (*0 GG ll−<) shifts current-account relation (30) rightward → higher relative private consumption and domestic appreciation (lower e).
- Mechanisms summarized:
  - Permanent retrenchment reduces future tax burden → raises lifetime private wealth → immediate consumption increase.
  - Sectoral labor reallocation increases productive use of labor → increases resources and mitigates current-account deterioration.
  - Money-demand effect typically dominates current-account effect in the short run → appreciation required to equilibrate money market.
- Output and timing:
  - Due to appreciation and expenditure-switching, domestic output decreases relative to foreign in short run (since * yy eθ− =).
  - If retrenchment temporary (*0 ll−< with *0 ll==), no consumption reallocation occurs because short-run nominal rigidities prevent private-sector absorption of labor; aggregate consumption unchanged.

### Long-run effects (Section III.C)
- Relative consumption follows a unit root: * * cc cc − = −  (standard result in this framework). Long-run change in relative consumption equals short-run change.
- Long-run net foreign assets reduced by retrenchment:
  - Reduced-form expression for long-run net foreign assets b: equation (31)
    - ii ll * 00 0002 * * 22 11 (1)( ) () 111 1 () (1)() 1(1) G gg PP LL nK bll LLLK nKn K gggg KK θ δθ θ δθδ δθ −− =+−+ + −− +−+ +− −−  (equation (31) as in source)
    - K are positive constants depending on θ, σ, δ, and initial labor shares.
  - Finding: reduction in domestic government employment has a negative effect on long-run net foreign assets (b decreases).
  - Intuition: higher relative consumption (short and long run) plus short-run relative output decline worsen long-run external financial position.
- Long-run relative output (equation (32)):
  - ** ()( ) 111 yy bcc n θδθ θθ − = − + − − − −  (equation (32) as in source)
  - Since b falls and *cc− rises following public employment retrenchment, long-run domestic output relative to foreign increases.
  - Interpretation: sectoral shift increases private employment in long run; higher relative consumption raises labor supply in private sector and long-run output.

### Effects when retrenchment finances higher non-wage government consumption (Section III.D)
- Government budget relation for permanent policies: ** 0 () GG ggHl l − = − −  (equation (33) as in source).
- Modified current-account/consumption relation: equation (34)
  - * * 00 0 000 11111 (1)() [()]() 1()1 G gg PP LL K eccHll LLL θ δ θδ δθθ θδ − + = + − + − − − + −   (equation (34) as in source)
  - The square-bracket term is the sum of a positive and a negative term → ambiguous overall effect on short-run relative consumption and exchange rate when retrenchment finances increased non-wage government consumption.
- Intuition for ambiguity:
  - Financing higher non-wage consumption removes tax-reduction wealth effect → agents do not gain wealth from retrenchment → tends to depress consumption until private sector absorbs retrenched workers in long run.
  - Increased government consumption stimulates demand and profits → increases private wealth and consumption.
- Role of initial public wage bill share (0 H):
  - A larger initial public wage bill share increases likelihood of a negative short-run effect on private consumption because a larger fraction of income derived from public employment amplifies the negative wealth effect of retrenchment.

### Conclusions and suggested extensions (Section IV)
- Main contributions:
  - Distinction between government spending for public employment and non-wage public consumption within a “new open economy macroeconomics” framework.
  - Shown that a reduction in domestic public employment used to reduce the tax burden implies an immediate increase in domestic consumption relative to foreign and an appreciation of the domestic exchange rate.
  - When retrenchment finances higher non-wage government consumption, effects are ambiguous and depend on initial share of private income from public employment.
- Suggested extensions:
  - Introduce government investment to analyze a wider range of policies.
  - Relax Ricardian-equivalence environment (allow government debt to have real effects) to study interaction of spending composition with financing alternatives (e.g., along lines of Ganelli, 2004).

*Source: _wp0504*

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

### _wp0504 - References..............................................................................................................

### I. INTRODUCTION — framing and motivation
- Government spending in macro models is typically introduced as non-wage government consumption of goods and services.
- Spending for public employment is an important government spending component, while non-wage public consumption tends to show the least significant dynamics.
- Paradox identified (Alesina and Perotti, 1995, p. 224): government non-wage consumption often receives the most attention despite limited dynamics.
- Objective: analyze effects of government spending composition in a two-country general equilibrium sticky-price model to resolve this paradox.

### Literature position and model choice
- Existing exceptions distinguishing public employment vs. non-wage consumption: Finn (1998) and Lane and Perotti (2003).
- Differences from Finn (1998):
  - Use of a two-country model in the “new open economy macroeconomics” strand rather than a closed-economy flexible-price RBC model.
  - Allows study of both long-run full-employment equilibrium and short-run demand-determined output due to nominal rigidities.
  - Enables analysis of effects on exchange rate and current account.
- References cited as conceptual anchors: Obstfeld and Rogoff, 1995, 1996; Lane, 2001; Lane and Ganelli, 2003; Lane and Perotti (2003).

### Key comparative policy experiments analyzed
- Policy A: Permanent reduction in domestic government employment used to finance a reduction in general taxation.
  - Immediate result: increase in domestic private consumption compared to foreign private consumption.
  - Mechanism: agents anticipate lower lifetime tax burden, including in new long-run equilibrium when public-sector employment reduction is compensated by increased private-sector employment.
  - Sectoral shift yields more productive use of domestic resources, easing current account pressures and allowing a more appreciated short-run exchange rate.
- Policy B: Reduction in public employment used to finance an increase in government non-wage consumption.
  - Macroeconomic effects are ambiguous.
  - Domestic agents do not gain from a lower tax burden, but benefit from expanded demand due to higher public consumption.
  - Analytical result: stronger pressure for reduction in domestic relative private consumption emerges when the initial share of private income derived from employment in the public sector is larger.

### Mechanisms emphasized
- Anticipation of lifetime tax changes drives private consumption responses.
- Sectoral employment shift affects productivity and external balances (current account, exchange rate).
- Composition of government spending (wages/public employment vs. non-wage consumption) alters both demand-side effects and distributional/incidence channels.

### Structure of the paper (as described)
- Next section: introduces the model.
- Section III: discusses effects of two macroeconomic policies with a simple graphical apparatus.
- Further sections (not provided here) presumably expand on comparisons with Finn (1998) and other literature.

*Source: _wp0504 - References..............................................................................................................*

### Section IV concludes.

### _wp0504 - Section IV concludes.

### Model setup and key mechanisms
- Two-country world (Home and Foreign) with continuum of infinitely-lived agents; total world population normalized to 1. Home agents on interval [0 n], foreign on [1]n,. A measure 1 of firms; n firms domestic, 1-n firms foreign.
- Agents derive utility from consumption of a composite good, leisure and real balances; key parameters and variables: β (discount factor), σ (intertemporal elasticity of consumption), χ and k (positive parameters), B (riskless real bond denominated in composite consumption good), t r (real interest rate), t M (nominal money balances at beginning of period t), C (private consumption index), P (price index), W (nominal wage), t Π (profit shares), t τ (lump-sum taxes).
- Private consumption index and price index: equations (4) and (5) with monopoly/monopolistic competition parameter 1θ>.
- Law of one price / PPP: P = E P∗ (E price of foreign currency in domestic currency).
- First-order conditions (representative domestic agent): Euler equation (6), money demand equation (7), labor-leisure trade-off (8).
- Production: labor-only production function Y = P t L (equation (9)); relationship between product price and nominal wage: equation (10).
- Government budget constraint (real terms): equation (11) — government finances non-wage consumption and government employment with lump-sum taxes and seignorage.

### Short-run nominal rigidities and log-linearization
- Short-run nominal rigidities: one period preset prices in producer currency (Obstfeld and Rogoff timing); domestic-currency prices of domestic goods and foreign-currency prices of foreign goods rigid in the period of the shock, adjust thereafter.
- Notation: tildes for short-run log-linearized variables, hats for long-run.
- Log-linearized system summarized in Table 1 (equations (13)–(24)): price index relation (13), Euler (14), short- and long-run money demand (15),(16), labor-leisure and labor market equilibrium (17),(18), demand and current-account relations (19)–(21), government budget (22), PPP (23), world goods market (24).
- Initial steady state: government spending for goods and net foreign assets zero for both countries; denoted with zero subscript (0GG BB====).

### Two equilibrium relationships and graphical solution
- Money-demand relationship (short-run relative consumption and exchange rate): equation (25)
  - 1 ()ecc σ =− −   (equation (25) as in source)
  - Interpretation: increase in domestic consumption relative to foreign raises money demand; requires domestic appreciation (decrease in e) to restore money-market equilibrium (“money demand” effect).
- Current-account relationship (consumption, exchange rate, and policy variables): equation (30)
  - ii * ** 00 000 1111 (1) ()()() 1()1 G gg PP LL K eccllgg LLL θ θδ δθθθ − =+−+−+ − −+−  ll * 1 () (1) gg δθ +− − . (equation (30) as in source)
  - Depends on policy variables (government employment and non-wage consumption); expresses a positive relationship between e and *cc− (worsened current account → depreciation).
- Graphical apparatus: downward-sloping (25) and upward-sloping (30) curves determine short-run equilibrium in (e, *cc−) space (Figure 1).

### Effects of a reduction in government employment (composition: reduced public employment + tax reduction)
- Short-run effects (Section III.B, Figure 2):
  - Asymmetric permanent decrease in domestic government employment relative to foreign ( *0 GG ll−< ) shifts equation (30) rightward → higher relative private consumption and domestic appreciation (lower e).
  - Mechanisms:
    - Permanent retrenchment reduces future tax burden → raises lifetime private wealth → immediate increase in consumption (rational expectations).
    - Sectoral labor reallocation from public to private increases productive use of labor → increases resources and mitigates current-account deterioration.
    - Money-demand effect dominates current-account effect in short-run for exchange rate determination → appreciation required to equilibrate money market.
  - Due to appreciation and expenditure-switching, domestic output decreases compared to foreign in short run (since * yy eθ− =).
- Key timing caveat:
  - If retrenchment temporary ( *0 ll−< with *0 ll== ), no consumption reallocation occurs because short-run nominal rigidities prevent private-sector absorption of labor; aggregate consumption unchanged.

### Long-run effects (Section III.C)
- Relative consumption follows a unit root: * * cc cc − = −  (standard result in this framework).
  - Hence long-run change in relative consumption equals short-run change.
- Long-run net foreign assets b expression (reduced form): equation (31)
  - ii ll * 00 0002 * * 22 11 (1)( ) () 111 1 () (1)() 1(1) G gg PP LL nK bll LLLK nKn K gggg KK θ δθ θ δθδ δθ −− =+−+ + −− +−+ +− −−  (equation (31) as in source)
  - K definitions: K positive constants depending on θ, σ, δ, and initial labor shares.
- Finding: reduction in domestic government employment has a negative effect on long-run net foreign assets (b decreases).
  - Intuition: higher relative consumption (short and long run) plus short-run relative output decline worsen long-run external financial position.
- Long-run relative output (equation (32)):
  - ** ()( ) 111 yy bcc n θδθ θθ − = − + − − − −  (equation (32) as in source)
  - Since b falls and *cc− rises following public employment retrenchment, long-run domestic output relative to foreign increases.
  - Interpretation: sectoral shift increases private employment in long run; higher relative consumption raises labor supply in private sector and long-run output.

### Effects of shifting spending composition: using retrenchment to finance non-wage government consumption (Section III.D)
- Government budget relation for permanent policies: ** 0 () GG ggHl l − = − −  (equation (33) as in source).
- Modified current-account/consumption relation: equation (34)
  - * * 00 0 000 11111 (1)() [()]() 1()1 G gg PP LL K eccHll LLL θ δ θδ δθθ θδ − + = + − + − − − + −   (equation (34) as in source)
  - The square-bracket term is the sum of a positive and a negative term → ambiguous overall effect on short-run relative consumption and exchange rate when retrenchment finances increased non-wage government consumption.
- Intuition for ambiguity:
  - Financing higher non-wage consumption removes tax-reduction wealth effect (agents do not benefit from lower tax burden) → net loss of labor income until private sector absorbs retrenched workers in long run → tends to depress consumption.
  - Increased government consumption stimulates demand and profits → increases private wealth and consumption.
- Role of initial public wage bill share (0 H):
  - A larger initial public wage bill share increases likelihood of leftward shift (negative effect on consumption) because larger fraction of income derived from public employment amplifies negative wealth effect of retrenchment.

### Conclusions and suggested extensions (Section IV)
- Main contributions:
  - Introduced distinction between government spending for public employment and non-wage public consumption within a “new open economy macroeconomics” framework.
  - Shown that a reduction in domestic public employment used to reduce the tax burden implies an immediate increase in domestic consumption relative to foreign and an appreciation of the domestic exchange rate.
  - When retrenchment finances higher non-wage government consumption, effects are ambiguous and depend on initial share of private income from public employment.
- Suggested extensions:
  - Introduce government investment to analyze a wider range of policies.
  - Relax Ricardian-equivalence environment (allow government debt to have real effects) to study interaction of spending composition with financing alternatives (e.g., along lines of Ganelli, 2004).

*Source: _wp0504 - Section IV concludes.*

### REFERENCES

### REFERENCES

### References list
- Alesina, Alberto, and Roberto Perotti, 1995, “Fiscal Expansions and Adjustments in OECD Countries,” Economic Policy, pp. 207–48.
- Finn, Mary G., 1998, “Cyclical Effects of Government Employment and Goods Purchases,” International Economic Review, Vol. 39, pp. 635–57.
- Ganelli, Giovanni, 2003, “Useful Government Spending, Direct Crowding-Out and Fiscal Policy Interdependence,” Journal of International Money and Finance, Vol. 22 (1), pp. 87–103.
- Ganelli, Giovanni, 2004, “The New Open Economy Macroeconomics of Government Debt,” forthcoming in the Journal of International Economics.
- Hau, Harald, 2000, “Exchange Rate Determination: the Role of Factor Price Rigidities and Nontradeables,” Journal of International Economics, Vol. 50 (2), pp. 421–47.
- Lane, Philip R., 2001, “The New Open Economy Macroeconomics: A Survey,” Journal of International Economics, Vol. 54 (2), pp. 235–66.
- Lane, Philip R., and Giovanni Ganelli, 2003,“Dynamic General Equilibrium Analysis: The Open Economy Dimension,” Elements in Dynamic Macroeconomic Analysis, ed. by S. Altug, J. Chadha, Lane Phillip, and C. Nolan (Cambridge: Cambridge University Press).
- Lane, Philip R., and Roberto Perotti, 2003, “The Importance of Composition of Fiscal Policy: Evidence from Different Exchange Rate Regimes,” Journal of Public Economics, Vol. 87, pp. 2253–79.
- Obstfeld, Maurice, and Kenneth Rogoff, 1996, ”Foundations of International Macroeconomics,” Chapter 10, Cambridge, Massachusetts: MIT Press.
- Obstfeld, Maurice, and Kenneth Rogoff, 1995 “Exchange Rate Dynamics Redux,” Journal of Political Economy, Vol. 103, pp. 624–60.

*Source: _wp0504 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp0504.pdf_