Week 8 · Alon Cohen

Aggregate Demand II: Applying the IS-LM Model

Aggregate Demand II: Applying the IS-LM Model

Part of: Macro-Economics II | Key concepts: IS-LM Equilibrium, Fiscal Policy, Monetary Policy, Crowding Out, IS Shocks, LM Shocks, Aggregate Demand Curve, AD-AS Model


What This Chapter Is About

Chapter 11 built the IS-LM model from scratch. Chapter 12 puts it to work. The goal is to use IS-LM as a policy and shock-analysis tool, and then to connect the IS-LM model back to the aggregate demand (AD) curve introduced in Chapter 10.

The three main tasks are:

  1. Use IS-LM to analyse fiscal and monetary policy — what happens to output YY and the interest rate rr when government spending, taxes, or the money supply change.
  2. Use IS-LM to analyse shocks — exogenous shifts in the goods market (IS shocks) or money market (LM shocks).
  3. Derive the AD curve from IS-LM — show why the AD curve is downward sloping and how policy shifts it.

A running theme is the interaction between monetary and fiscal policy: in the real world, the central bank may react to fiscal changes, and this dramatically affects outcomes.


Part 1: IS-LM Equilibrium — Quick Review

The IS-LM model delivers a unique short-run equilibrium (Y∗,r∗)(Y^*, r^*) that simultaneously clears the goods market and the money market:

IS: Y=C(Y−T)+I(r)+G⟺goods market\text{IS: } Y = C(Y-T) + I(r) + G \quad \Longleftrightarrow \quad \text{goods market}
LM: MˉPˉ=L(r,Y)⟺money market\text{LM: } \frac{\bar{M}}{\bar{P}} = L(r, Y) \quad \Longleftrightarrow \quad \text{money market}

Graphically, the equilibrium is the intersection of the IS curve (downward sloping in rr-YY space) and the LM curve (upward sloping). Any policy or shock that shifts either curve changes the equilibrium (Y∗,r∗)(Y^*, r^*).

IS vs LM — Who Controls What?
Curve Moved by Slopes
IS Fiscal policy (GG, TT); investment expectations Downward
LM Monetary policy (MM); money demand shocks Upward

Fiscal policy shifts IS; monetary policy shifts LM. This is the key operating principle for all the policy analysis below.


Part 2: Fiscal Policy Analysis

An Increase in Government Purchases (ΔG>0\Delta G > 0)

When government raises spending by ΔG\Delta G, the IS curve shifts rightward.

Why? At any given interest rate rr, higher GG raises planned expenditure PE=C+I+GPE = C + I + G. For the goods market to remain in equilibrium, output YY must rise to match — so the IS curve shifts right.

How far does IS shift? The horizontal shift equals the Keynesian cross multiplier effect:

ΔY∣r fixed=11−MPC⋅ΔG\Delta Y\bigg|_{r \text{ fixed}} = \frac{1}{1 - MPC} \cdot \Delta G

But the IS-LM model modifies this: once output starts to rise, money demand L(r,Y)L(r, Y) increases too (more income → more transactions → more need for cash). Since money supply is unchanged, rr must rise to equilibrate the money market. The higher rr then crowds out investment.

The full story:

Step What happens
1 ↑G\uparrow G → IS shifts right
2 At old rr, output would rise by ΔG1−MPC\frac{\Delta G}{1-MPC} (Keynesian multiplier)
3 Higher YY → ↑\uparrow money demand → ↑r\uparrow r (LM curve unchanged)
4 Higher rr → ↓I\downarrow I (crowding out) → partially offsets the initial ↑Y\uparrow Y
5 Net result: ↑Y\uparrow Y and ↑r\uparrow r, but ΔY<ΔG1−MPC\Delta Y < \frac{\Delta G}{1-MPC}
Crowding Out: The Key Caveat to the Multiplier

In the Keynesian cross (from Chapter 11), ΔY/ΔG=1/(1−MPC)\Delta Y / \Delta G = 1/(1-MPC). This assumes rr is fixed. In IS-LM, the rise in YY feeds back into the money market, raising rr, which reduces investment. This crowding-out effect means the actual multiplier in IS-LM is smaller than the Keynesian cross multiplier. How much smaller depends on the slopes of IS and LM.

Two Forces Acting Simultaneously
  • Multiplier effect: ΔG\Delta G raises income, which raises consumption, which raises income further…
  • Crowding-out effect: the rise in rr reduces investment, partially offsetting the multiplier

The larger the crowding-out relative to the multiplier, the less effective fiscal policy is at raising output.

A Tax Cut (ΔT<0\Delta T < 0)

A tax cut increases disposable income (Y−T)(Y - T), raising consumption and shifting the IS curve rightward — but by less than an equivalent increase in GG.

Why smaller IS shift? A ΔG\Delta G of €1 raises expenditure by €1 directly. A ΔT\Delta T of €-1 raises disposable income by €1, but households save (1−MPC)(1 - MPC) of that — only MPCMPC enters spending immediately. So:

IS shift from ΔG:11−MPC⋅ΔG\text{IS shift from } \Delta G: \quad \frac{1}{1-MPC} \cdot \Delta G
IS shift from ΔT:MPC1−MPC⋅∣ΔT∣(smaller)\text{IS shift from } \Delta T: \quad \frac{MPC}{1-MPC} \cdot |\Delta T| \quad \text{(smaller)}
Worked Example: Tax Multiplier in IS-LM (MPC = 0.75)

Suppose MPC = 0.75 and a tax cut of ΔT=−100\Delta T = -100 ends up raising YY by 200 (given crowding-out from the IS-LM interaction).

Step A — Keynesian tax multiplier (without IS-LM feedback): −MPC1−MPC=−0.750.25=−3\frac{-MPC}{1-MPC} = \frac{-0.75}{0.25} = -3 If no crowding-out, ΔT=−100\Delta T = -100 should raise YY by 3×100=3003 \times 100 = 300. But YY only rises by 200 — the gap (100100) is the crowding-out from higher rr.

Step B — Decompose ΔC\Delta C and ΔI\Delta I: ΔC=MPC⋅(ΔY−ΔT)=0.75×(200−(−100))=0.75×300=225\Delta C = MPC \cdot (\Delta Y - \Delta T) = 0.75 \times (200 - (-100)) = 0.75 \times 300 = 225 ΔI=ΔY−ΔC=200−225=−25(crowded out)\Delta I = \Delta Y - \Delta C = 200 - 225 = -25 \quad \text{(crowded out)}

Step C — Changes in savings: ΔSg=−ΔT=+(−100)=−100(public savings fall)\Delta S_g = -\Delta T = +(-100) = -100 \quad \text{(public savings fall)} ΔSp=MPS⋅(ΔY−ΔT)=0.25×300=75(private savings rise)\Delta S_p = MPS \cdot (\Delta Y - \Delta T) = 0.25 \times 300 = 75 \quad \text{(private savings rise)} ΔStotal=−100+75=−25=ΔI✓\Delta S_{total} = -100 + 75 = -25 = \Delta I \checkmark

This confirms: both statements can be true — private savings rise, but public savings fall by more, so total savings fall and investment is crowded out.


Part 3: Monetary Policy Analysis

An Increase in Money Supply (ΔM>0\Delta M > 0)

Expansionary monetary policy shifts the LM curve right (or equivalently, downward):

Why? An increase in MM raises real balances M/PM/P (since PP is fixed in the short run). At any income level YY, the money market now has excess supply — money supply exceeds demand. To eliminate excess supply, rr must fall to stimulate money demand. The LM curve shifts right (= same YY now consistent with a lower rr).

The full transmission:

Step What happens
1 ↑M\uparrow M → LM shifts right/down
2 At the initial r1r_1, money supply exceeds demand
3 People buy bonds → bond prices rise → rr falls
4 Lower rr → ↑I\uparrow I → higher planned expenditure → ↑Y\uparrow Y
5 Higher YY → ↑\uparrow money demand → rr rises back somewhat
6 New equilibrium: lower r∗r^*, higher Y∗Y^*
A Common Mistake in Monetary Policy Diagrams

After expansionary monetary policy, the interest rate does NOT simply fall to the level implied by the money market alone (where LM crosses the yy-axis). The IS-LM model shows a subtlety:

  • When rr falls, II rises → YY rises → money demand rises → this pushes rr back up from its trough
  • The final equilibrium interest rate r∗r^* is higher than the initial fall to r2r_2 in the money market alone — but still lower than r1r_1

Graphically: LM shifts down; we move along the IS curve to a new intersection. The final r∗r^* is between r1r_1 and the "pure money market" lower bound.


Part 4: Interaction Between Monetary and Fiscal Policy

So far, fiscal and monetary policy have been treated as independent. In reality, the central bank may react to fiscal policy — or vice versa. This interaction dramatically changes outcomes.

Three Scenarios: How the Fed Responds to a Fiscal Expansion (ΔG>0\Delta G > 0)

Congress raises GG → IS shifts right. What does the Fed do?

Fed Response Mechanism Effect on YY Effect on rr Key Insight
1. Hold MM constant LM unchanged ↑Y\uparrow Y (moderate) ↑r\uparrow r Standard IS-LM result; some crowding-out
2. Hold rr constant Fed increases MM, LM shifts right ↑Y\uparrow Y (large, = Keynesian cross result) Unchanged No crowding-out; fiscal multiplier maximised
3. Hold YY constant Fed reduces MM, LM shifts left Unchanged ↑r\uparrow r (sharply) Complete crowding-out; ΔG=ΔI=ΔS\Delta G = \Delta I = \Delta S

Response 1: Hold MM Constant (Benchmark IS-LM)

The standard result: output rises moderately, interest rate rises. Crowding-out is partial. This is the normal IS-LM textbook scenario.

Response 2: Hold rr Constant (Keynesian Model)

The Fed accommodates fiscal expansion by increasing MM to prevent the interest rate from rising. Since rr doesn't change, there is no crowding-out of investment. The fiscal multiplier operates at full Keynesian force: ΔY=11−MPC⋅ΔG\Delta Y = \frac{1}{1-MPC} \cdot \Delta G.

Decomposing the income change:

  • ΔY=ΔG1−MPC\Delta Y = \frac{\Delta G}{1-MPC}
  • ΔC=MPC1−MPC⋅ΔG\Delta C = \frac{MPC}{1-MPC} \cdot \Delta G
  • ΔS=ΔI=0\Delta S = \Delta I = 0 (since rr is unchanged)

Response 3: Hold YY Constant (Classical Long-Run Equivalent)

The Fed tightens money to prevent any output change. The IS shift right is exactly offset by the LM shift left. Output stays at Yˉ\bar{Y} but rr rises sharply.

This is the long-run classical result in disguise: in the long run, YY is determined by supply-side factors, so fiscal expansion only raises rr and crowds out investment one-for-one: ΔG=−ΔI\Delta G = -\Delta I.

Which Response Is "Right"?

None is universally correct — it depends on economic conditions. If the economy is in a deep recession (below potential Yˉ\bar{Y}), Response 2 maximises stimulus. If the economy is at full employment, Response 3 captures the classical insight that stimulus only displaces private investment.


Part 5: Shocks in the IS-LM Model

Shocks are exogenous movements in the IS or LM curve — not caused by deliberate policy.

IS Shocks

IS shocks arise from exogenous changes in the demand for goods and services.

Examples:

  • Stock market boom → higher household wealth → ↑C\uparrow C → IS shifts right
  • Stock market crash → lower wealth → ↓C\downarrow C → IS shifts left
  • Surge in business optimism → ↑I\uparrow I → IS shifts right
  • Fall in consumer confidence (e.g. after 9/11) → ↓C\downarrow C and ↓I\downarrow I → IS shifts left

LM Shocks

LM shocks arise from exogenous changes in the demand for money, at a given YY and rr.

Examples:

  • Wave of credit card fraud → consumers demand more cash → ↑Ld\uparrow L^d → LM shifts left (= upward)
  • More ATMs or online banking → less cash needed → ↓Ld\downarrow L^d → LM shifts right (= downward)
Analysing Two Shocks with IS-LM

Shock A: Housing market crash (IS shock) Households' wealth falls → consumption ↓\downarrow → IS shifts left. Results: Y↓Y \downarrow, r↓r \downarrow, C↓C \downarrow (lower income and wealth), I↑I \uparrow (lower rr partially offsets), u↑u \uparrow (Okun's law: lower YY → higher unemployment).

Shock B: Increase in money demand (LM shock) Identity theft makes people hold more cash → ↑Ld\uparrow L^d → LM shifts left (up). Results: Y↓Y \downarrow, r↑r \uparrow, C↓C \downarrow (lower income), I↓I \downarrow (higher rr), u↑u \uparrow.

Key contrast: an IS shock moves rr and YY in the same direction (both fall). An LM shock moves rr and YY in opposite directions (rr rises while YY falls).


Case Study: The U.S. Recession of 2001

The 2001 recession is a textbook illustration of multiple simultaneous IS shocks, followed by both fiscal and monetary policy responses.

The Shocks (All Shifted IS Leftward)

1. Stock market decline (2000–2001)

The dot-com bubble burst. The S&P 500 lost nearly half its value. Lower household wealth reduced consumption CC and lower equity valuations reduced business investment II. IS shifted left.

2. September 11, 2001

The terrorist attacks sharply increased uncertainty and reduced consumer and business confidence. Spending on durables and investment collapsed. IS shifted further left.

3. Corporate accounting scandals (Enron, WorldCom)

Revelations of massive fraud destroyed trust in public companies, reduced stock prices, and discouraged both equity investment and business investment. IS shifted left again.

Macroeconomic outcome:

  • 2.1 million jobs lost
  • Unemployment rose from 3.9% to 5.8%
  • GDP growth slowed to 0.8% (vs 3.9% average during 1994–2000)

Policy Responses

Fiscal policy (shifting IS rightward):

  • Bush tax cuts (2001 and 2003): ↓T\downarrow T → ↑\uparrow disposable income → ↑C\uparrow C
  • Increased government spending: ↑G\uparrow G directly
  • Airline industry bailout; NYC reconstruction; Afghanistan war spending

Monetary policy (shifting LM rightward):

  • Fed dramatically cut interest rates: the 3-month T-bill rate fell from ~6% to ~1%
  • Lower rr stimulated investment and spending
Why IS-LM Explains the 2001 Recession Perfectly

Multiple simultaneous IS shocks explain the recession. Policy responded by trying to shift IS and LM back to the right. The combination of fiscal stimulus and aggressive monetary easing (LM shifts right) ultimately stabilised output — illustrating precisely the policy interaction mechanics of Part 4.


Part 6: Deriving the Aggregate Demand Curve from IS-LM

So far in Chapter 12, prices PP have been assumed fixed. But what happens when PP changes?

The Mechanism

The IS-LM model operates with real money balances M/PM/P. If PP rises:

P↑⇒MP↓⇒LM shifts left/up⇒r↑⇒I↓⇒Y↓P \uparrow \Rightarrow \frac{M}{P} \downarrow \Rightarrow \text{LM shifts left/up} \Rightarrow r \uparrow \Rightarrow I \downarrow \Rightarrow Y \downarrow

If PP falls:

P↓⇒MP↑⇒LM shifts right/down⇒r↓⇒I↑⇒Y↑P \downarrow \Rightarrow \frac{M}{P} \uparrow \Rightarrow \text{LM shifts right/down} \Rightarrow r \downarrow \Rightarrow I \uparrow \Rightarrow Y \uparrow

The AD Curve

By varying PP and recording the IS-LM equilibrium YY at each PP, we trace out the aggregate demand (AD) curve:

AD curve: negative relationship between P and Y, derived from IS-LM\boxed{AD \text{ curve}: \text{ negative relationship between } P \text{ and } Y, \text{ derived from IS-LM}}
Why Is the AD Curve Downward Sloping?

The underlying logic is the interest rate effect (working through investment II): P↑⇒M/P↓⇒r↑⇒I↓⇒Y↓P \uparrow \Rightarrow M/P \downarrow \Rightarrow r \uparrow \Rightarrow I \downarrow \Rightarrow Y \downarrow

A secondary channel is the wealth effect: higher PP erodes the real value of financial assets → lower household wealth → lower CC → lower YY.

Note: the AD curve is NOT downward sloping for the same reasons as a regular demand curve (substitution/income effects). It slopes down because higher prices tighten real money supply, raising rr and reducing investment.

Policy Shifts the AD Curve

Any policy that shifts IS or LM at a given price level PP also shifts the entire AD curve:

Policy/Shock Effect on IS/LM Effect on AD Curve
↑G\uparrow G or ↓T\downarrow T (expansionary fiscal) IS shifts right AD shifts right
↓G\downarrow G or ↑T\uparrow T (contractionary fiscal) IS shifts left AD shifts left
↑M\uparrow M (expansionary monetary) LM shifts right AD shifts right
↓M\downarrow M (contractionary monetary) LM shifts left AD shifts left
Positive IS shock (↑\uparrow investment optimism) IS shifts right AD shifts right
Negative IS shock (↓\downarrow confidence) IS shifts left AD shifts left
AD vs IS/LM: What Moves What
  • Movement along the AD curve: caused by PP changing (LM shifts, IS-LM intersection moves)
  • Shift of the AD curve: caused by policy or shocks (IS or LM shifts at a given PP, changing YY for every PP)

Part 7: Short-Run vs Long-Run — IS Shocks and the AD-AS Model

Once we have the AD curve, we can reconnect IS-LM to the AD-AS framework from Chapter 10 to trace out both short-run and long-run effects.

SR Effect of a Negative IS Shock

A negative IS shock (e.g. stock market crash) shifts IS left and AD left. In the short run:

  • YY falls below natural rate Yˉ\bar{Y}
  • PP falls slightly (or stays fixed — below SRAS)
  • rr falls

The economy is in a recessionary gap.

The Transition to Long-Run Equilibrium

Over time, when Y<YˉY < \bar{Y}, the price level gradually falls (or inflation slows). This triggers:

  1. P↓⇒M/P↑⇒P \downarrow \Rightarrow M/P \uparrow \Rightarrow LM shifts right/down
  2. Lower rr → ↑I\uparrow I → YY rises back toward Yˉ\bar{Y}
  3. SRAS shifts down (lower wages and input prices)
  4. This process continues until Y=YˉY = \bar{Y} in long-run equilibrium
Long-run: Y returns to Yˉ, but P is permanently lower\text{Long-run: } Y \text{ returns to } \bar{Y}, \text{ but } P \text{ is permanently lower}
Monetary Expansion: SR and LR Effects

Short run: ↑M\uparrow M → LM shifts right → r↓r \downarrow, Y↑Y \uparrow above Yˉ\bar{Y}, PP unchanged.

Transition: Y>YˉY > \bar{Y} → PP gradually rises → M/PM/P falls → LM shifts back left → rr rises back up.

Long run: PP has risen proportionally to the rise in MM. All real variables — YY, rr, II, CC — return to their initial values. Only PP (the nominal variable) is permanently higher.

Conclusion: money is neutral in the long run. The short-run effect on YY is temporary; the long-run effect is purely on the price level.

IS-LM and AD-AS: Connecting the Frameworks

Horizon Price Output Key Mechanism
Short run Fixed (sticky) Demand-determined via IS-LM IS-LM intersection
Transition Gradually adjusts Moves back to Yˉ\bar{Y} SRAS shifts + LM endogenous response
Long run Fully flexible Supply-determined at Yˉ\bar{Y} LRAS vertical; only PP changes

The force that moves the economy from short run to long run is the gradual adjustment of prices:

Y>Yˉ⇒P↑ over timeY > \bar{Y} \Rightarrow P \uparrow \text{ over time}
Y<Yˉ⇒P↓ over timeY < \bar{Y} \Rightarrow P \downarrow \text{ over time}
Y=Yˉ⇒P stableY = \bar{Y} \Rightarrow P \text{ stable}
Why Policy Can Shorten the Recession

After a negative IS shock, the economy would eventually self-correct as prices fall and LM drifts right. But this can take years. Policymakers can speed up recovery by using expansionary fiscal or monetary policy to shift IS and/or LM rightward, restoring YY to Yˉ\bar{Y} immediately without waiting for price adjustment.

The trade-off: restoring YY via policy keeps PP from falling (or even raises it), whereas waiting for self-correction lowers PP permanently. If policymakers want to stabilise output and minimise the budget deficit, combining fiscal + monetary policy (Response 2 logic from Part 4) minimises the required deficit increase.


Summary Table: IS-LM Policy and Shock Analysis

Scenario IS Effect LM Effect Net ΔY\Delta Y Net Δr\Delta r
↑G\uparrow G (fiscal expansion) Right Unchanged ++ ++ (crowding-out)
↓T\downarrow T (tax cut) Right (smaller) Unchanged ++ ++ (crowding-out, less)
↑M\uparrow M (monetary expansion) Unchanged Right/Down ++ −-
Negative IS shock Left Unchanged −- −-
Negative LM shock (↑\uparrow money demand) Unchanged Left/Up −- ++
Combined expansionary policy Right Right/Down ++++ Ambiguous

Summary

  1. IS-LM as a policy tool: fiscal policy shifts the IS curve; monetary policy shifts the LM curve. Both raise YY via different channels — fiscal via ↑\uparrow expenditure (↑G\uparrow G / ↓T\downarrow T); monetary via ↓r\downarrow r → ↑I\uparrow I.

  2. Crowding out: fiscal expansion raises YY, but also raises rr, which reduces investment. The final ΔY\Delta Y in IS-LM is smaller than the Keynesian cross multiplier alone predicts, because the money market feedback partially reverses the stimulus.

  3. Fed response matters: if the Fed holds rr constant (accommodates fiscal policy with ↑M\uparrow M), fiscal multiplier is maximised. If Fed holds YY constant, fiscal policy only raises rr with zero output effect (complete crowding-out).

  4. Shocks: IS shocks (wealth, confidence) move rr and YY in the same direction. LM shocks (money demand shifts) move rr and YY in opposite directions.

  5. 2001 recession: multiple simultaneous IS shocks (dot-com crash, 9/11, accounting scandals) → IS/AD shifted left → recession. Fiscal and monetary policy responded by shifting IS and LM right to restore demand.

  6. AD curve: derived by varying PP through the IS-LM model. Downward sloping because higher PP reduces M/PM/P, raises rr, lowers II, and lowers YY. Policy and shocks shift the AD curve.

  7. Short-run vs long-run: after a shock, YY deviates from Yˉ\bar{Y} in the short run. Over time, gradual price adjustment shifts SRAS and endogenously moves LM, returning the economy to Yˉ\bar{Y}. Money is neutral in the long run.