Skip to content
TheChatArchiveThe Chat Archive
ClaudeClaude Haiku 4.5· anthropicHugging Face Frontier Reasoning Traces

Derive the IS and LM curves mathematically showing investment-saving and liquidity…

  1. User

    Derive the IS and LM curves mathematically showing investment-saving and liquidity preference-money supply relationships. Solve for equilibrium interest rate and output.

  2. Claude Assistant

    This is a comprehensive request to derive the IS-LM model mathematically. I'll work through this systematically, showing all the derivations clearly.

    Let me structure this as:

    1. The IS curve derivation (Investment-Saving equilibrium)
    2. The LM curve derivation (Liquidity preference-Money supply equilibrium)
    3. Solving for the general equilibrium
    4. Provide numerical examples

    IS CURVE DERIVATION

    The IS curve comes from the goods market equilibrium condition:

    National Income Accounting Identity: Y = C + I + G + NX

    For a closed economy, we ignore NX, so: Y = C + I + G

    This is the fundamental national accounting identity.

    In Equilibrium (Goods Market): S + T ≡ I + G (where we can derive this from national accounting)

    Actually, let me be more precise. The income-expenditure approach says: Y = C + I + G

    Rearranging: Y - C = I + G Y - C - T + T = I + G (adding and subtracting taxes) (Y - T - C) + T = I + G S + T = I + G

    So in equilibrium: S = I + G - T

    Or more commonly: S = I + (G - T)

    Where:

    • S = Private saving
    • I = Investment
    • G = Government spending
    • T = Taxes

    More specifically:

    • Saving: S = Y - C - T (disposable income minus consumption)
    • Consumption: C = C₀ + c(Y - T) where C₀ is autonomous consumption and c is MPC
    • Investment: I = I₀ - bi (depends on interest rate i)
    • G and T are exogenous

    Deriving the IS Curve:

    Starting with: Y = C + I + G

    Substituting consumption and investment functions:

    Y = C₀ + c(Y - T) + I₀ - bi + G

    Expanding and collecting terms: Y - cY = C₀ - cT + I₀ - bi + G

    Y(1 - c) = C₀ + I₀ + G - cT - bi

    Solving for output: Y = [C₀ + I₀ + G - cT - bi]/(1 - c)

    This can be rewritten as: Y = [1/(1-c)][C₀ + I₀ + G - cT - bi]

    The multiplier 1/(1-c) shows how changes in autonomous spending affect equilibrium income, while the -bi term captures how interest rates inversely affect investment and thus output. The IS curve demonstrates how interest rates and output move inversely.

    For the LM curve, I need to work from money market equilibrium. Money demand consists of two parts: transactions demand tied to income levels, and speculative demand that responds to interest rates. Using Keynes's liquidity preference framework, money demand is expressed as L(Y, i) = kY - hi, where k captures how income changes affect money needs and h reflects the inverse relationship between interest rates and speculative holdings. Both parameters are positive—higher income increases transaction demand while higher rates reduce the desire to hold money for speculation.

    The central bank sets the money supply exogenously at M₀. Equilibrium occurs when money supply equals money demand: M₀ = kY - hi. Rearranging this equation yields the LM curve: i = (k/h)Y - (1/h)M₀. This reveals the opposite relationship from the IS curve—output and interest rates move together on the LM curve.

    The economy reaches equilibrium where both curves intersect, meaning both the goods market and money market clear simultaneously. Setting the IS and LM equations equal allows me to solve for the equilibrium levels of output and interest rates.

    Combining the two conditions and solving for Y*, I get:

    Y = [hkA₀ + kbM₀]/(k²b + h)*

    This equilibrium output depends on both autonomous spending (A₀) and the money supply (M₀). To find the equilibrium interest rate, I substitute Y* back into the LM equation to determine i*. i = [A₀(k²b + h) - (hkA₀ + kbM₀)]/[kb(k²b + h)]

    i = [k²bA₀ + hA₀ - hkA₀ - kbM₀]/[kb(k²b + h)]

    i = [k²bA₀ + hA₀(1-k) - kbM₀]/[kb(k²b + h)]

    The algebra is getting messy with this notation. I should switch to clearer variable names—using α for the marginal propensity to consume instead of k will make the expressions more transparent.

    Setting up the IS curve, I need to work through the goods market equilibrium. Output equals aggregate demand, which combines consumption (with autonomous spending C₀ and marginal propensity α applied to disposable income), investment that responds negatively to interest rates with sensitivity b, and exogenous government spending and taxes. The LM curve then relates money supply to money demand, which depends on income through parameter m and on interest rates through parameter h. Y - hi

    hi = mY - M

    LM: i = (m/h)Y - M/h

    Setting the IS and LM curves equal to find equilibrium:

    A/b - Y/(kb) = (m/h)Y - M/h

    Rearranging to isolate Y:

    A/b + M/h = Y[m/h + 1/(kb)]

    A/b + M/h = Y[(mkb + h)/(hkb)]

    Y = [k(hA + bM)]/(mkb + h)*

    The equilibrium output depends on both fiscal policy (through A) and monetary policy (through M), with their relative influence determined by the structural parameters of the model.

    For the interest rate, I substitute Y* back into the IS equation:

    i* = A/b - Y*/(kb)

    where Y* = [k(hA + bM)]/(mkb + h)

    This gives:

    i* = A/b - [(hA + bM)]/[b(mkb + h)]

    i* = [A(mkb + h) - (hA + bM)]/[b(mkb + h)]

    i* = [Amkb + Ah - hA - bM]/[b(mkb + h)]

    i* = [Amkb - bM]/[b(mkb + h)]

    i* = [mb(kA - M/k)]/[b(mkb + h)]

    The algebra gets messy this way, so I'll use the LM equation instead. From i = (m/h)Y - M/h and substituting the equilibrium output:

    i* = [mk(hA + bM)]/(h(mkb + h)) - M/h

    i* = [mkhA + mkbM - Mmkb - Mh]/(h(mkb + h))

    i* = [mkhA - Mh]/(h(mkb + h))

    i = (mkA - M)/(mkb + h)*

    Now checking this against the IS equation, I'm getting an inconsistency when I try to verify. Let me reconsider the IS relationship—specifically whether i = A/b - Y/(kb) is actually correct. Going back to Y = kA - kbi, I can rearrange to get i in terms of Y, but I need to be careful about how the multiplier k enters the expression.

    Actually, reworking the IS curve from the consumption and investment components: Y = C₀ + α(Y - T) + I₀ - bi + G. Collecting Y terms and defining A = C₀ + I₀ + G - αT gives Y(1 - α) = A - bi, which means i = A/b - Y(1 - α)/b. So the IS curve should have (1 - α) in the denominator, not k. I made an error in my earlier derivation.

    But if I substitute k = 1/(1 - α), then (1 - α) = 1/k, and the IS equation becomes i = A/b - Y/(kb). That matches what I had before, so the IS curve is correct.

    Now I'm checking the equilibrium by setting IS equal to LM. With IS: i = A/b - Y/(kb) and LM: i = (m/h)Y - M/h, I can solve for Y by finding a common denominator on the right side of the equilibrium equation. The LCD of h and kb is hkb, so (m/h) becomes mkb/(hkb) and 1/(kb) becomes h/(hkb). This gives me Y[(mkb + h)/(hkb)] on the right side after combining terms.

    Solving for Y yields Y* = [k(hA + bM)]/(mkb + h). Using the LM equation to find the equilibrium interest rate, I substitute this Y* value and simplify to get i* = (mkA - M)/(mkb + h). I can verify this is correct by plugging Y* back into the IS equation—the algebra works out to the same result. b + h)] - (hA + bM)/[b(mkb + h)]

    i = [A(mkb + h) - (hA + bM)]/[b(mkb + h)]

    i = [Amkb + Ah - hA - bM]/[b(mkb + h)]

    i = [Amkb - bM]/[b(mkb + h)]

    i = (Amk - bM)/(mkb + h)

    This doesn't align with what I derived from the LM equation. From that approach I got (mkA - M)/(mkb + h), but here I'm getting (Amk - bM)/(mkb + h). These are only equivalent if b equals 1. Let me trace back through the LM derivation to verify where the discrepancy originates. khA - Mh]/[h(mkb + h)]

    Simplifying by separating terms: i = [mkA]/(mkb + h) - M/(mkb + h) = (mkA - M)/(mkb + h)

    From the IS equation, I get: i = [Amkb - b²M]/[b(mkb + h)] = [Amk]/(mkb + h) - [bM]/(mkb + h) = (Amk - bM)/(mkb + h)

    The two expressions don't match—LM gives (mkA - M)/(mkb + h) while IS gives (Amk - bM)/(mkb + h). Since mkA and Amk are identical, the numerators differ by a factor of b in the second term, indicating an error in one of my derivations.

    Rechecking the IS curve more carefully: starting from Y = A/(1-α) - bi/(1-α), the relationship between i and Y is i = (1-α)(A - Y)/b. Using k = 1/(1-α), this becomes i = (A - Y)/(kb). I had incorrectly separated this into individual terms when I should have kept it as a single fraction. The correct IS curve is i = (A - Y)/(kb), not the form I was using before.

    Now setting the IS and LM curves equal to solve for equilibrium. From IS: i = (A - Y)/(kb), and from LM: i = (m/h)Y - M/h. Equating these and solving through the algebra yields Y* = (hA + kbM)/(mkb + h). The equilibrium interest rate follows from substituting back into either equation.

    For i*, I substitute the equilibrium output into the IS curve. After simplifying the numerator—which reduces to kb(Am - M) after canceling terms—I get i* = (Am - M)/(mkb + h). I can verify this using the LM equation to confirm both expressions are consistent.

    The equilibrium values are Y* = (hA + kbM)/(mkb + h) and i* = (mA - M)/(mkb + h). A quick dimensional check confirms these make sense: h carries units of inverse interest rate, A represents income, and the denominator mkb + h has consistent units throughout. From M = mY - hi:

    • M is in money units
    • Y is in income units
    • m has units of (money/income)
    • i is in interest rate units
    • h has units of (money/interest)

    From IS: i = (A - Y)/(kb)

    • b represents ΔY/Δi, giving it units of (income/interest)
    • k is the multiplier ΔY/ΔA, which is dimensionless
    • kb therefore has dimensions (income/interest)
    • (A-Y)/(kb) yields interest units ✓

    For the equilibrium solution Y* = (hA + kbM)/(mkb + h), the numerator combines hA and kbM, both yielding (money·income/interest). The denominator mkb + h also produces (money/interest) units, making the ratio dimensionally consistent for income.

    Similarly, i* = (mA - M)/(mkb + h) has a numerator in money units and denominator in (money/interest), confirming the result is in interest rate units.

    The key relationships are the IS curve i = (A - Y)/(kb) and LM curve i = (m/h)Y - M/h, which determine equilibrium output Y* = (hA + kbM)/(mkb + h) and equilibrium interest rate i* = (mA - M)/(mkb + h).

    To make this concrete, I'll work through a numerical example with specific parameter values: autonomous consumption of 100, MPC of 0.75, taxes of 200, autonomous investment of 200, government spending of 300, investment sensitivity to interest of 50, income elasticity of money demand at 0.2, interest elasticity at 100, and money supply of 1000. First, autonomous spending equals 100 + 200 + 300 - 0.75(200) = 450, and the multiplier is 1/(1-0.75) = 4. Plugging into the equilibrium formulas gives the actual values for output and interest rates. The parameters seem unrealistic given the negative rate, so I'll adjust the money supply upward to M = 2000.

    With this revision, the equilibrium output becomes Y* = (100(450) + 4(50)(2000))/140 = 445000/140 ≈ 3178.57, and the interest rate calculation shifts to i* = (90 - 2000)/140, which still yields a negative result but reflects the impact of the increased money supply on the system.

    The persistent negative rates suggest the money demand sensitivity parameter m itself needs recalibration. Increasing m to 0.5 changes the denominator to 200, producing Y* = 2225 and i* = -8.875—far too negative. This indicates the money demand specification may be fundamentally misaligned with the model's structure.

    For positive equilibrium interest rates, I need mA > M. With A = 450 and m = 0.2, this gives mA = 90, so M must stay below 90. Let me restart with parameters that ensure realistic outcomes: C₀ = 100, α = 0.75, T = 100, I₀ = 100, and G = 150.

    I'll set b = 20 for investment sensitivity to interest rates, m = 0.2 for money's income elasticity, h = 50 for asset demand, and M = 50 for the money supply.

    First, I calculate A = 100 + 100 + 150 - 0.75(100) = 275. Then k = 1/(1-0.75) = 4. The equilibrium output follows from Y* = (h·A + M)/(mkb + h).

    Working through the components: h·A = 13,750, kb = 80, so mkb + h = 66. This gives Y* ≈ 268.94. For the interest rate, i* = (m·A - M)/(mkb + h) = 5/66 ≈ 7.58%.

    I verify both conditions hold: the IS curve yields i ≈ 0.0758 when Y = 268.94, and the LM curve should confirm this same rate at equilibrium output. M = mY - hi

    These represent the same equilibrium relationship in different forms.

    The IS curve shows how interest rates and output move inversely—higher rates discourage investment and reduce aggregate demand. Its slope depends on investment sensitivity and the multiplier strength. Autonomous spending shifts it rightward. The LM curve captures the positive link between output and rates: as production rises, money demand increases, pushing rates up to clear the money market at a fixed supply level. The slope reflects how responsive money demand is to income versus interest rate changes, and monetary expansion shifts it right.

    Policy changes produce distinct effects on equilibrium. Fiscal stimulus raises both output and interest rates because the income boost increases money demand. Monetary expansion works differently—it lowers rates while raising output, the standard transmission mechanism. The relative potency of each policy hinges on parameter values, particularly how sensitive investment and money demand are to interest rates.

    IS-LM Model: Mathematical Derivation and Equilibrium Solution

    I. DERIVING THE IS CURVE (Goods Market Equilibrium)

    National Income Accounting Identity

    Y=C+I+GY = C + I + G

    Behavioral Functions

    Consumption Function: C=C0+α(Y−T)C = C_0 + \alpha(Y - T)

    where α\alpha = MPC (0 < α < 1), C0C_0 = autonomous consumption

    Investment Function: I=I0−biI = I_0 - bi

    where I0I_0 = autonomous investment, bb = investment sensitivity, ii = interest rate

    Equilibrium Condition

    Substituting into income identity: Y=C0+α(Y−T)+I0−bi+GY = C_0 + \alpha(Y - T) + I_0 - bi + G

    Y−αY=C0+I0+G−αT−biY - \alpha Y = C_0 + I_0 + G - \alpha T - bi

    Y(1−α)=A0−biY(1 - \alpha) = A_0 - bi

    where A0=C0+I0+G−αTA_0 = C_0 + I_0 + G - \alpha T (autonomous spending)

    Solving for IS Curve

    Y=A0−bi1−αY = \frac{A_0 - bi}{1 - \alpha}

    Let k=11−αk = \frac{1}{1-\alpha} (spending multiplier)

    Y=k(A0−bi)=kA0−kbiY = k(A_0 - bi) = kA_0 - kbi

    IS CURVE: i=A0−Ykb\boxed{i = \frac{A_0 - Y}{kb}}

    Alternative form: i=A0kb−Ykbi = \frac{A_0}{kb} - \frac{Y}{kb}


    II. DERIVING THE LM CURVE (Money Market Equilibrium)

    Keynes's Liquidity Preference

    Money demand depends on transactions (income) and speculative (interest-rate) motives:

    Md=mY−hiM^d = mY - hi

    where:

    • mm = income sensitivity (transactions demand)
    • hh = interest rate sensitivity (asset demand)
    • Both m,h>0m, h > 0

    Money Market Equilibrium

    Money supply is exogenously fixed: Ms=MM^s = M

    At equilibrium: Ms=MdM^s = M^d

    M=mY−hiM = mY - hi

    hi=mY−Mhi = mY - M

    LM CURVE: i=mhY−Mh\boxed{i = \frac{m}{h}Y - \frac{M}{h}}

    Rearranged: M=mY−hiM = mY - hi


    III. SOLVING FOR GENERAL EQUILIBRIUM

    Setting IS = LM

    A0−Ykb=mhY−Mh\frac{A_0 - Y}{kb} = \frac{m}{h}Y - \frac{M}{h}

    Multiply both sides by kbkb:

    A0−Y=mkbhY−kbhMA_0 - Y = \frac{mkb}{h}Y - \frac{kb}{h}M

    A0+kbhM=mkbhY+YA_0 + \frac{kb}{h}M = \frac{mkb}{h}Y + Y

    A0+kbhM=Y(mkb+hh)A_0 + \frac{kb}{h}M = Y\left(\frac{mkb + h}{h}\right)

    Equilibrium Output

    Y∗=hA0+kbMmkb+h\boxed{Y^* = \frac{hA_0 + kbM}{mkb + h}}

    Equilibrium Interest Rate

    Substitute Y∗Y^* back into LM:

    i∗=mh⋅hA0+kbMmkb+h−Mhi^* = \frac{m}{h} \cdot \frac{hA_0 + kbM}{mkb + h} - \frac{M}{h}

    i∗=m(hA0+kbM)−M(mkb+h)h(mkb+h)i^* = \frac{m(hA_0 + kbM) - M(mkb + h)}{h(mkb + h)}

    i∗=mhA0+mkbM−Mmkb−Mhh(mkb+h)i^* = \frac{mhA_0 + mkbM - Mmkb - Mh}{h(mkb + h)}

    i∗=mA0−Mmkb+h\boxed{i^* = \frac{mA_0 - M}{mkb + h}}


    IV. NUMERICAL EXAMPLE

    Given Parameters:

    ParameterValueMeaning
    C0C_0100Autonomous consumption
    α\alpha0.75Marginal propensity to consume
    I0I_0100Autonomous investment
    GG150Government spending
    TT100Taxes
    bb20Investment elasticity
    mm0.2Money demand/income sensitivity
    hh50Money demand/interest elasticity
    MM50Money supply

    Calculations

    Step 1: Autonomous spending A0=100+100+150−0.75(100)=275A_0 = 100 + 100 + 150 - 0.75(100) = 275

    Step 2: Multiplier k=11−0.75=4k = \frac{1}{1-0.75} = 4

    Step 3: Equilibrium output Y∗=50(275)+4(20)(50)0.2(4)(20)+50=13,750+4,00016+50=17,75066=268.94Y^* = \frac{50(275) + 4(20)(50)}{0.2(4)(20) + 50} = \frac{13,750 + 4,000}{16 + 50} = \frac{17,750}{66} = 268.94

    Step 4: Equilibrium interest rate i^* = \frac{0.2(275) - 50}{16 + 50} = \frac{55 - 50}{66} = \frac{5}{66} = 0.0758 \text{ (7.58%)}

    Verification

    IS Check: i=275−268.9480=0.0758i = \frac{275 - 268.94}{80} = 0.0758 ✓

    LM Check: i=0.250(268.94)−1=1.076−1=0.076i = \frac{0.2}{50}(268.94) - 1 = 1.076 - 1 = 0.076 ✓


    V. ECONOMIC INTERPRETATIONS

    AspectIS CurveLM Curve
    RelationshipNegative: ↑i → ↓YPositive: ↑Y → ↑i
    MechanismHigher rates reduce investmentHigher income increases money demand
    Slope−1kb-\frac{1}{kb} (flatter with large bb or kk)mh\frac{m}{h} (steeper with large mm)
    ShiftsRight when A0A_0 ↑Right when MM ↑

    Comparative Statics

    Fiscal Expansion (↑A0↑A_0): ∂Y∗∂A0=hmkb+h>0,∂i∗∂A0=mmkb+h>0\frac{\partial Y^*}{\partial A_0} = \frac{h}{mkb+h} > 0, \quad \frac{\partial i^*}{\partial A_0} = \frac{m}{mkb+h} > 0 → Crowding out: expenditure increases but higher rates reduce investment

    Monetary Expansion (↑M↑M): ∂Y∗∂M=kbmkb+h>0,∂i∗∂M=−1mkb+h<0\frac{\partial Y^*}{\partial M} = \frac{kb}{mkb+h} > 0, \quad \frac{\partial i^*}{\partial M} = -\frac{1}{mkb+h} < 0 → Lower rates stimulate investment and output


    VI. KEY PROPERTIES

    1. Stability: System is stable if IS slope > LM slope: 1kb>mh\frac{1}{kb} > \frac{m}{h}

    2. Policy Effectiveness:

      • Fiscal policy more effective when hh is large (money demand interest-sensitive) or bb is small (investment interest-inelastic)
      • Monetary policy more effective when bb is large or hh is small (liquidity trap risk when h→∞h → ∞)
    3. Multiplier: Equilibrium multiplier <k< k due to interest rate feedback

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.