254 lines
8.9 KiB
Markdown
254 lines
8.9 KiB
Markdown
# Connecting Theory to Data: Twin Deficits Analysis
|
|
|
|
## Overview
|
|
This document connects the theoretical concepts in the Monetary Policy Interactive Diagram to the empirical analysis from Problem Set 1 (Twin Deficits Hypothesis, 1960-2024 US data).
|
|
|
|
## National Accounting Identity
|
|
|
|
### Theoretical Framework
|
|
The fundamental identity that ties everything together:
|
|
|
|
$$Y = C + I + G + CA$$
|
|
|
|
Where:
|
|
- **Y** = GDP/National Income
|
|
- **C** = Consumption
|
|
- **I** = Investment
|
|
- **G** = Government purchases
|
|
- **CA** = Current Account (= Exports - Imports)
|
|
|
|
### Savings Identity
|
|
Rearranging the national income identity:
|
|
|
|
$$Y - C - G = I + CA$$
|
|
|
|
Since National Savings $S = Y - C - G$, we get:
|
|
|
|
$$S = I + CA$$
|
|
|
|
Breaking down savings into private and government components:
|
|
- $S_p = Y - T - C$ (Private savings)
|
|
- $S_g = T - G$ (Government savings/budget balance)
|
|
|
|
Therefore:
|
|
|
|
$$\boxed{CA = (S_p - I) + S_g = (S_p - I) + (T - G)}$$
|
|
|
|
**This is the core equation linking government budget balance to the current account.**
|
|
|
|
## Empirical Evidence: US Data (1960-2024)
|
|
|
|
### Finding 1: Twin Deficits Correlation Changed
|
|
|
|
**Before 1990 (1960-1989)**:
|
|
- Correlation: **r = 0.82** (very strong positive)
|
|
- Both CA and government budget relatively balanced
|
|
- When government ran deficits, current account followed strongly
|
|
|
|
**After 1990 (1990-2024)**:
|
|
- Correlation: **r = 0.53** (moderate positive)
|
|
- Both deficits became structurally larger and persistent
|
|
- Twin deficits still evident, but relationship more complex
|
|
|
|
### Finding 2: Private Savings Collapse
|
|
|
|
| Metric | Before 1990 | After 1990 | Change |
|
|
|--------|-------------|------------|--------|
|
|
| **Private Savings/GDP** | 8.05% | 4.67% | **-42%** |
|
|
| **Investment/GDP** | 18.22% | 17.70% | -3% |
|
|
| **S-I Gap** | -10.16% | -13.03% | -2.87 pp |
|
|
|
|
**Key Insight**: The dramatic decline in private savings fundamentally changed the US economy's relationship with foreign capital.
|
|
|
|
## Theoretical Explanation
|
|
|
|
### Why the Correlation Weakened But Twin Deficits Persisted
|
|
|
|
Looking at the identity: $CA = (S_p - I) + (T - G)$
|
|
|
|
**Before 1990**:
|
|
- $(S_p - I)$ was relatively stable at around -10%
|
|
- Changes in $(T - G)$ directly translated to changes in CA
|
|
- This created the strong correlation (r = 0.82)
|
|
|
|
**After 1990**:
|
|
- $(S_p - I)$ became larger and more variable (dropped to -13%)
|
|
- The current account now depends on TWO moving parts:
|
|
1. Government balance $(T - G)$
|
|
2. Private sector balance $(S_p - I)$
|
|
- This reduced the simple correlation but both deficits grew larger
|
|
|
|
### Mathematical Representation
|
|
|
|
**Before 1990** (stable private sector):
|
|
$$\Delta CA \approx \Delta(T - G)$$
|
|
Because $\Delta(S_p - I) \approx 0$
|
|
|
|
**After 1990** (variable private sector):
|
|
$$\Delta CA = \Delta(S_p - I) + \Delta(T - G)$$
|
|
Both terms matter!
|
|
|
|
## Connection to Monetary Policy
|
|
|
|
### Interest Rate Channel
|
|
From the interactive diagram, when central bank raises rates:
|
|
|
|
1. **Direct Effect**: $R \uparrow$ → $I \downarrow$ (investment falls)
|
|
2. **If private savings don't rise enough**: $(S_p - I)$ increases
|
|
3. **From identity**: If $(S_p - I) \uparrow$ and $(T-G)$ constant → $CA \uparrow$
|
|
|
|
This is the **expenditure-switching** effect of monetary policy.
|
|
|
|
### Exchange Rate Channel
|
|
Monetary tightening also works through UIP:
|
|
|
|
1. $R_{domestic} \uparrow$ → Currency appreciates ($E \downarrow$)
|
|
2. Exports become less competitive
|
|
3. $CA \downarrow$ (deteriorates)
|
|
|
|
These two channels can work in **opposite directions**!
|
|
|
|
### Net Effect in US Case (Post-1990)
|
|
|
|
The data shows:
|
|
- Persistent current account deficits despite varying interest rates
|
|
- This suggests the **private savings collapse** dominated
|
|
- Even when government improved its balance (late 1990s), CA remained negative
|
|
- Because $(S_p - I)$ gap was too large
|
|
|
|
## Policy Implications
|
|
|
|
### 1. Fiscal Policy Alone Insufficient
|
|
Simply reducing government deficits won't fix current account if:
|
|
$$S_p - I < 0 \text{ and large}$$
|
|
|
|
**Example**: Late 1990s US had government **surplus** but still ran CA deficit because private savings were too low relative to investment.
|
|
|
|
### 2. Monetary Policy Trade-offs
|
|
Tight monetary policy to fight inflation creates tensions:
|
|
- Higher $R$ reduces $I$ (good for CA via S-I gap)
|
|
- But higher $R$ appreciates currency (bad for CA via trade)
|
|
- Net effect ambiguous
|
|
|
|
### 3. Structural Factors Matter
|
|
The private savings rate depends on:
|
|
- **Demographics**: Aging population saves less
|
|
- **Financial development**: Easy credit reduces precautionary savings
|
|
- **Social safety nets**: Strong pensions reduce need to save
|
|
- **Income inequality**: High inequality can depress aggregate savings
|
|
|
|
## Using the Interactive Diagram
|
|
|
|
### Scenario: Replicate 1990s US Economy
|
|
|
|
**Settings**:
|
|
1. Set Government Balance: Moving toward surplus
|
|
2. Set Private Savings: Low (represented by high consumption)
|
|
3. Set Investment: Moderate to high
|
|
4. Observe: CA can still be negative despite government surplus
|
|
|
|
### Scenario: 2022-2023 Inflation Response
|
|
|
|
**Settings**:
|
|
1. Set Inflation: 4-5% (above target)
|
|
2. Set Output Gap: +2% (overheating)
|
|
3. Taylor Rule prescribes: $R = 2\% + 1.5(3\%) + 0.5(2\%) = 7.5\%$
|
|
4. Observe effects on:
|
|
- Investment (falls sharply)
|
|
- Exchange rate (appreciates)
|
|
- Net exports (deteriorate)
|
|
|
|
## Data Visualization Integration
|
|
|
|
### Recommended Addition to Interactive Tool
|
|
|
|
Create a time-series plotter showing:
|
|
1. **Government Balance** (T - G)
|
|
2. **Private Balance** (S_p - I)
|
|
3. **Current Account** (CA)
|
|
4. **Vertical line at 1990** to show structural break
|
|
|
|
This would show empirically how:
|
|
$$CA_t = (S_p - I)_t + (T - G)_t$$
|
|
|
|
Students could see all three variables moving together (or not).
|
|
|
|
## Equations Summary
|
|
|
|
### Core Identities
|
|
1. **National Income**: $Y = C + I + G + CA$
|
|
2. **Savings**: $S = S_p + S_g = I + CA$
|
|
3. **Current Account**: $CA = (S_p - I) + (T - G)$
|
|
|
|
### Monetary Policy
|
|
4. **Money Market**: $\frac{M^s}{P} = L(R, Y)$
|
|
5. **Taylor Rule**: $R = R^* + 1.5(\pi - \pi^*) + 0.5(y - y^*)$
|
|
6. **UIP**: $R_{CHF} = R_{EUR} + \frac{E^e - E}{E}$
|
|
|
|
### Relationships
|
|
7. **Investment**: $I = I(R, Y)$, where $\frac{\partial I}{\partial R} < 0$
|
|
8. **Net Exports**: $NX = NX(E, Y, Y^*)$, where $\frac{\partial NX}{\partial E} > 0$
|
|
9. **Consumption**: $C = C(Y - T, R, Wealth)$
|
|
|
|
## Discussion Questions
|
|
|
|
### 1. Twin Deficits Hypothesis
|
|
**Q**: If the correlation weakened after 1990, does this mean the twin deficits hypothesis failed?
|
|
|
|
**A**: No! The hypothesis is based on the **identity** $CA = (S_p - I) + (T - G)$, which is always true. The correlation weakened because $(S_p - I)$ became a larger and more variable component, not because the identity broke down.
|
|
|
|
### 2. Ricardian Equivalence
|
|
**Q**: Could Ricardian equivalence explain the changing relationship?
|
|
|
|
**A**: Ricardian equivalence predicts $\Delta S_p = -\Delta(T-G)$ (private savings offset government dissaving). If true, $(S_p - I) + (T-G)$ would be constant and CA wouldn't change with government balance.
|
|
|
|
**Evidence**: US data shows this did NOT happen. Private savings fell independently of government balance, suggesting Ricardian equivalence doesn't hold empirically.
|
|
|
|
### 3. Foreign Capital Dependence
|
|
**Q**: What does persistent CA deficit mean for the US?
|
|
|
|
**A**: From $CA = S - I$, a negative CA means $S < I$, so:
|
|
- US invests more than it saves domestically
|
|
- Must attract foreign capital to finance the gap
|
|
- This makes the US a **net debtor** to the rest of the world
|
|
- Sustainable only if foreigners willing to hold US assets
|
|
|
|
### 4. Exchange Rate Regime
|
|
**Q**: Would a different exchange rate regime change the relationship?
|
|
|
|
**A**: Under **fixed exchange rates**, the identity still holds but adjustment mechanism differs:
|
|
- Can't use monetary policy independently (UIP ties R to foreign rate)
|
|
- Current account imbalances must adjust through price/wage changes
|
|
- This is slower and potentially more painful
|
|
|
|
Under **flexible rates** (US case):
|
|
- Exchange rate adjusts to help balance current account
|
|
- But as data shows, adjustment can be slow and incomplete
|
|
|
|
## Conclusion
|
|
|
|
The interactive monetary policy diagram and the twin deficits empirical analysis are deeply connected:
|
|
|
|
1. **Identities** provide the theoretical foundation
|
|
2. **Policy rules** (Taylor, UIP) show how variables adjust
|
|
3. **Empirical data** reveals which channels dominate in practice
|
|
|
|
**Key lesson**: Theory gives us the framework, but magnitudes and relative importance of different channels require empirical investigation. The US experience shows that structural changes (private savings collapse) can fundamentally alter economic relationships even when identities remain valid.
|
|
|
|
## Further Reading
|
|
|
|
**Textbook Chapters**:
|
|
- Open economy macroeconomics
|
|
- Fiscal and monetary policy interactions
|
|
- International capital flows
|
|
|
|
**Papers**:
|
|
- Chinn & Prasad (2003): "Medium-term determinants of current accounts"
|
|
- Obstfeld & Rogoff (2005): "Global current account imbalances"
|
|
- Bernanke (2005): "The global saving glut" (speech)
|
|
|
|
**Data Sources**:
|
|
- FRED (Federal Reserve Economic Data): All US national accounts
|
|
- IMF Balance of Payments Statistics
|
|
- OECD Economic Outlook Database
|