Crowding Out: When Government Borrowing Displaces Private Investment

Crowding out is a fundamental but often overlooked economic mechanism: when governments borrow heavily, they increase demand for loanable funds, pushing up interest rates. Higher interest rates make borrowing expensive for businesses, reducing private investment. Government spending has "crowded out" private investment. Understanding crowding out is essential for evaluating whether government spending is truly stimulative (increasing total investment and growth) or merely redistributive (replacing private investment with public investment). This analysis examines the mechanisms, empirical evidence, and implications for fiscal policy design.

Quick definition: Crowding out occurs when government borrowing increases demand for loanable funds, raising interest rates and reducing private business investment. The result: public spending displaces private spending dollar-for-dollar (or partially), reducing net economic stimulus.

Key Takeaways

• Crowding out is strong in full employment (30-50% of government borrowing crowds out private investment) and weak in recessions (near-zero crowding out)
• Empirical evidence from multiple countries suggests 30-50% crowding out in normal economic times
• Interest rate increase from government borrowing is typically 15-20 basis points per $100 billion borrowing increase
• Long-term crowding out: Sustained high debt reduces capital accumulation, lowering productivity growth and long-term living standards
• Optimal fiscal policy runs large deficits during recessions (no crowding out) and small deficits during expansions (avoiding crowding out)

The Loanable Funds Market: Where Borrowers and Savers Meet

The loanable funds market is the financial market where savers lend to borrowers (businesses, governments, consumers). Interest rates equilibrate supply and demand, determining how much is lent and at what rate.

Supply of Loanable Funds: Where Funds Come From

Loanable funds come from multiple sources:

Household savings:

• Savings accounts and money market funds
• Retirement accounts (401k, IRA, pensions)
• Direct purchases of bonds
• Household saving rate: Approximately 3-8% of income (varies with interest rates and confidence)

Business savings:

• Corporate retained earnings
• Cash flow from operations
• Depreciation allowances (available for reinvestment)
• Share of loanable funds: Approximately 40-50%

Government savings/borrowing:

• Government budget surplus (borrowing negative; adding to loanable funds)
• Government budget deficit (borrowing positive; reducing loanable funds)
• Share: Depends on fiscal policy (deficit reduces available loanable funds)

Foreign investor savings:

• Foreign citizens and governments holding dollar assets
• Foreign investors lending to U.S. businesses and government
• Share: Approximately 20-30% of total U.S. loanable funds (varies with exchange rates)

Financial intermediaries:

• Banks creating credit (depositors' funds lent to borrowers)
• Money creation through credit expansion
• Share: Significant, but varies with monetary policy and credit conditions

Total U.S. loanable funds supply: Approximately $3-4 trillion per year (varies with economic conditions and interest rates)

Demand for Loanable Funds: Who Borrows

Borrowers demand loanable funds for various purposes:

Business investment:

• Capital equipment, buildings, research
• Expected return must exceed borrowing cost to justify investment
• Most sensitive to interest rates (higher rates reduce investment)

Government borrowing:

• To finance budget deficits
• Borrowing amount determined by spending less tax revenue
• Not sensitive to interest rates (government borrows desired amount regardless of rate)

Consumer borrowing:

• Mortgages for home purchases
• Car loans
• Student loans
• Credit card borrowing
• Moderately sensitive to interest rates

Total demand: Typically $3-4 trillion per year (equilibrium with supply)

Market Equilibrium: Interest Rate Adjustment

The interest rate adjusts until quantity of loanable funds demanded equals quantity supplied.

Example equilibrium:

• Supply of loanable funds: $3.5 trillion at various interest rates
• Demand for loanable funds: $3.5 trillion at current interest rate
• Equilibrium interest rate: 4%
• Equilibrium quantity: $3.5 trillion borrowed

If demand shifts (government borrows more), interest rates must rise until quantity supplied equals new quantity demanded.

How Crowding Out Works: The Mechanism

When government borrowing increases, it shifts the demand curve for loanable funds outward. If the supply of loanable funds is constrained (as it is in full employment with high savings), interest rates must rise.

Scenario 1: Full Employment (Crowding Out Strong)

Initial equilibrium (before government borrowing increase):

• Loanable funds supply: $3.5 trillion per year
• Government borrowing: $600 billion
• Business investment: $2.1 trillion
• Consumer borrowing: $800 billion
• Interest rate: 4%
• All three categories sum to loanable funds supplied

Government increases deficit by $200 billion (new borrowing: $800 billion instead of $600 billion):

• This shifts government demand outward by $200B
• Total demand for loanable funds: $3.7 trillion
• But supply of loanable funds: Only $3.5 trillion (unchanged)

Market adjustment:

• Interest rates must rise to reduce quantity demanded
• Higher rates make borrowing more expensive
• Business investment falls (some projects no longer profitable at higher rates)
• Consumer borrowing falls (mortgages and auto loans become less affordable)
• Interest rate rises to 4.2% or 4.3% (approximately 20-30 basis points)

New equilibrium:

• Loanable funds supplied: $3.5 trillion (unchanged)
• Government borrowing: $800 billion (as desired)
• Business investment: $2.05 trillion (fell by $50 billion)
• Consumer borrowing: $750 billion (fell by $50 billion)
• Crowding out: $100 billion of the $200 billion government borrowing crowds out private borrowing

Result: 50% crowding out—half of government borrowing displaces private investment.

Scenario 2: Recession (Crowding Out Weak or Zero)

Initial equilibrium (during recession):

• Loanable funds supply: $4.0 trillion (businesses saving rather than investing; consumers saving)
• Government borrowing: $600 billion
• Business investment: $1.5 trillion (depressed by low demand expectations)
• Consumer borrowing: $900 billion
• Interest rate: 2% (low due to excess loanable funds)

Government increases deficit by $200 billion:

• Total demand: $3.7 trillion
• Supply of loanable funds: $4.0 trillion (actually higher because recession increases savings)

Market adjustment:

• Loanable funds are abundant (supply exceeds demand even with government borrowing)
• Interest rates fall slightly (to 1.8% or 1.9%) due to excess supply
• Business investment remains unchanged (rate fell, not rose; investment might even increase)
• Consumer borrowing unchanged (rate fell)

New equilibrium:

• Government borrowing: $800 billion (as desired)
• Business investment: $1.5 trillion (unchanged; no crowding out)
• Consumer borrowing: $900 billion (unchanged)
• Interest rate: 1.8% (actually falls)

Result: 0% crowding out—government borrowing does not displace private investment because loanable funds are abundant.

Graphical Illustration of Crowding Out

Before Government Borrowing Increase

Interest Rate (%)
    |
  6 |        Supply of Loanable Funds
    |       /
  5 |      /
    |     /    Equilibrium A
  4 |    /     (4%, Q1 quantity)
    |   /
  3 |  / \
    | /   \
  2 |/     \ Demand (Private Borrowing)
    |       \
  1 |        \___
    |_____________________ Quantity of Loanable Funds
              Q1

Equilibrium A: Interest rate 4%, quantity Q1. All borrowing is private (no government deficit in this scenario).

After Government Borrowing Increase

Interest Rate (%)
    |
  6 |        Supply
    |       /
  5 |      /
    |     /         Equilibrium B
  4 |    /          (5.5%, Q2 quantity)
    |   /
  3 |  / \
    | /   \
  2 |/     \ Demand with Government Borrowing
    |       \
  1 |        \___
    |_____________________ Quantity of Loanable Funds
              Q1  Q2

Equilibrium B: Interest rate rises to 5.5%, quantity Q2. Government borrowing is Q2-Q1. But private borrowing fell because of higher rates. This is crowding out.

When Does Crowding Out Occur? Economic Conditions Matter

The strength of crowding out depends critically on economic conditions and the elasticity of loanable funds supply.

Crowding Out Is Strong When:

1. Economy at full employment:

• All workers employed, factories operating at high capacity
• Limited idle resources available
• Loanable funds supply is limited
• Government borrowing must displace private borrowing
• Empirical evidence: 40-50% crowding out

2. Limited supply of loanable funds:

• Few domestic savers (low savings rate)
• Limited foreign investment inflows
• Restrictive monetary policy (central bank tight credit)
• Interest rates must rise significantly for government to borrow

3. Closed economy (limited international capital flows):

• Capital cannot flow in from abroad
• Foreign saving doesn't supplement domestic loanable funds
• Government borrowing demand must be met by domestic savers
• U.S. advantage: Open capital markets allow foreign investment; other countries less fortunate

Crowding Out Is Weak When:

1. Recession with idle resources:

• Unemployment elevated, factories running below capacity
• Unused savings available (households and businesses accumulating cash)
• Interest rates low (excess loanable funds relative to demand)
• Government borrowing doesn't need to outbid private borrowing
• Empirical evidence: 0-10% crowding out

2. Abundant supply of loanable funds:

• High savings rate (households saving more)
• Foreign capital inflows (investors lending to government)
• Expansionary monetary policy (central bank increasing credit supply)
• Interest rates can remain low even with government borrowing

3. Open economy with global capital flows:

• Foreign savers supply loanable funds
• When U.S. government borrows, not just domestic savers compete
• Foreign investors can substitute into U.S. Treasury bonds
• This reduces pressure on domestic interest rates

Empirical Evidence: How Much Does Crowding Out Actually Occur?

Economic research attempts to measure crowding out by observing how much private investment falls when government borrowing increases.

Consensus Estimates

In normal times (full employment/expansion):

• Crowding out: 30-50%
• For every $1 of government borrowing, private investment falls by $0.30-0.50
• Interest rate increase: 15-20 basis points per $100 billion borrowing
• Mechanism: Higher interest rates reduce business investment return

In recessions:

• Crowding out: 0-10% (near zero)
• For every $1 of government borrowing, private investment falls by $0-0.10
• Interest rate change: 0 basis points or negative (rates might fall)
• Mechanism: Loanable funds abundant; idle savings substitute

Research Sources and Studies

Federal Reserve research (2010-2020):

• Staff analysis suggests 30-40% crowding out in normal times
• Much lower crowding out in recession (COVID-19 analysis showed minimal crowding out 2020-2021 despite $3+ trillion borrowing)

Academic research:

• Salop (1974): 50% crowding out estimate
• Cebula (1979): 25-75% depending on model specification
• Elmendorf and Reifschneider (2002): 30-50% crowding out
• Literature varies widely, but consensus around 30-50% in normal times

Variation by country and context:

• U.S. experiences less crowding out than smaller countries (global capital access)
• Countries with capital controls experience stronger crowding out
• Fixed exchange rate regimes: Higher crowding out
• Floating exchange rate regimes: Lower crowding out (currency appreciation attracts capital)

Why Estimates Vary

Crowding out estimates vary for several reasons:

1. Time period matters: Different eras have different capital mobility and savings rates
2. Model specification: Different econometric approaches yield different estimates
3. Definition of crowding out: Some studies measure displacement of investment, others of interest rates
4. Reverse causation: Weak private investment might cause government to borrow more for stimulus (not crowding out causing weak investment)

Long-Term Crowding Out: Effects on Growth and Productivity

Beyond immediate displacement of investment, sustained high government debt creates long-term crowding out that affects growth rates.

The Long-Term Mechanism

Mechanism:

1. High government debt requires continuous borrowing to refinance
2. Large government borrowing keeps interest rates elevated
3. High interest rates reduce private business investment return
4. Private capital accumulation slows
5. Less capital per worker reduces productivity growth
6. Long-term growth rate falls

Example:

• U.S. debt-to-GDP rises from 100% to 150%
• Government borrowing increases from 3% to 5% of loanable funds
• Interest rates rise from 4% to 5%
• Business investment return requirements rise; some projects no longer profitable
• Capital accumulation falls from 3% to 2% per year
• Productivity growth falls from 2% to 1.5% per year
• Long-term living standard growth falls 20%

Empirical Evidence on Debt and Growth

Correlation evidence:

• Countries with very high debt-to-GDP (150%+) tend to grow slower (average 1-2% annually)
• Countries with moderate debt-to-GDP (50-80%) grow faster (3-4% annually)
• Countries with low debt (below 40%) show variable growth rates

Causation question: Does high debt cause slow growth, or does slow growth cause high debt?

Research findings:

• Likely both directions, but evidence suggests debt does constrain growth
• Studies using instrumental variables suggest causal effect: each 10 percentage point increase in debt-to-GDP reduces long-term growth by approximately 0.1-0.2 percentage points
• This is modest but real: sustained over decades, compounds into significant living standard differences

U.S. situation:

• Current debt-to-GDP: 122% (as of 2023)
• Historical average: 40-50% (pre-1980s)
• Risk level: Not yet in danger zone (150%+) but elevated
• If debt continues rising to 150-180%, growth could be noticeably constrained

The Stimulus Paradox: Crowding Out Reduces Stimulus Effectiveness

A key insight: Government spending that crowds out is less stimulative than spending that doesn't crowd out.

Example: Highway Spending During Recession vs. Expansion

Recession scenario (no crowding out):

• Government spends $100 billion on highway construction
• This employs construction workers, equipment manufacturers, engineers
• Workers earn wages and spend on food, housing, entertainment
• Restaurants, retailers, landlords profit from increased demand
• Total stimulus effect: $100 billion direct spending × multiplier of 1.5 = $150 billion GDP growth
• Private investment unchanged (loanable funds abundant, rates falling)
• Total investment increase: $100 billion government + $0 private = $100 billion
• Crowding out: 0%

Expansion scenario (strong crowding out):

• Government spends $100 billion on highway construction
• This crowds out approximately $40-50 billion of private business investment
• Government hires construction workers (same as recession)
• But private businesses don't expand factories/equipment as much (can't afford high interest rates)
• Employment gain is lower (fewer jobs from reduced private investment)
• Total stimulus effect: Much weaker (possibly zero if crowding out is 100%)
• Total investment increase: $100 billion government - $40 billion private = $60 billion net
• Crowding out: 40-50%

Implication: Stimulus works in recessions (no crowding out) but not in expansions (strong crowding out). This explains why fiscal stimulus appears effective in 2008-2009 (recession, low crowding out) but ineffective in 2022-2023 (expansion, high crowding out).

Numerical Example: U.S. Deficit and Crowding Out Calculation

Case 1: Government Increases Deficit by $200 Billion During Expansion

Baseline conditions (normal expansion, full employment):

• Supply of loanable funds: $3.5 trillion per year
• Government borrowing: $600 billion (peacetime deficit)
• Private business investment: $2.1 trillion
• Consumer/other borrowing: $800 billion
• Interest rate: 4%
• Total borrowing: $3.5 trillion (equilibrium)

Government increases spending by $200 billion (not offset by tax increases):

• New government borrowing: $800 billion (increase of $200B)
• Supply of loanable funds: $3.5 trillion (unchanged; savings unchanged, monetary policy unchanged)

Market adjustment:

• Total demand for loanable funds: $3.7 trillion
• Supply: $3.5 trillion
• Shortage of $200 billion at 4% rates
• Rates rise to clear market

Assume supply elasticity of 0.5 (for every 1% rate increase, supply increases 0.5%):

• Rate must rise to $4.28% (approximately 28 basis points)
• At 4.28%, private business investment falls by approximately $80 billion (demand elasticity -0.4)
• Consumer borrowing falls by approximately $20 billion
• Total private borrowing falls: $100 billion
• Crowding out: $100B / $200B = 50%

New equilibrium:

• Government borrowing: $800 billion
• Private business investment: $2.02 trillion (fell $80B)
• Consumer borrowing: $780 billion (fell $20B)
• Interest rate: 4.28%
• Total borrowing: $3.58 trillion (equilibrium with supply at higher rate)

Result:

• Government borrowing increased $200B (as desired)
• Private investment fell $100B (50% crowding out)
• Net increase in total investment: $100B
• Growth stimulus: Moderate, not the full $200B

Case 2: Same $200 Billion Increase During Recession

Recession conditions (idle capacity, high unemployment):

• Supply of loanable funds: $4.2 trillion per year (elevated; businesses and households saving)
• Government borrowing: $600 billion
• Private business investment: $1.5 trillion (depressed by low demand)
• Consumer/other borrowing: $1.1 trillion
• Interest rate: 2%
• Total borrowing: $3.2 trillion (supply exceeds demand, rates low)

Government increases borrowing by $200 billion:

• New government borrowing: $800 billion
• Supply of loanable funds: $4.2 trillion (even higher in recession; more savings)

Market adjustment:

• Total demand for loanable funds: $3.4 trillion
• Supply: $4.2 trillion
• Surplus of $800 billion at 2% rates
• Rates remain low or fall

New equilibrium:

• Government borrowing: $800 billion (as desired)
• Private business investment: $1.5 trillion (unchanged; rates fell or flat)
• Consumer borrowing: $1.1 trillion (unchanged)
• Interest rate: 1.5% (falls due to excess supply)
• Total borrowing: $3.4 trillion

Result:

• Government borrowing increased $200B
• Private investment unchanged ($0 crowding out)
• Net increase in total investment: $200B
• Growth stimulus: Strong, $200B multiplied by multiplier (1.5-2.0)

The Paradox Explained: Why Stimulus Works in Recessions But Not Expansions

The difference between these scenarios explains the key insight:

Recessions: Abundant idle loanable funds mean government borrowing doesn't compete with private borrowing. Deficit spending provides stimulus without crowding out. Stimulus multiplier is strong (1.5-2.0).

Expansions: Limited loanable funds mean government borrowing competes directly with private borrowing. Deficit spending crowds out private investment dollar-for-dollar (or partially). Stimulus multiplier is weak (0.5-1.0).

This is why economists recommend:

• Large deficits during recessions (stimulus works; crowding out minimal)
• Small deficits during expansions (avoid crowding out; let private investment drive growth)

Optimal Fiscal Policy Implication: Counter-Cyclical Deficits

Understanding crowding out suggests optimal fiscal policy is counter-cyclical (deficits when economy weak, surpluses when economy strong).

Optimal Policy Rule

During recessions (unemployment elevated, growth weak):

• Run large deficits (5-10% of GDP)
• Crowding out minimal
• Stimulus multiplier high (1.5-2.0)
• Stimulus effective at reducing unemployment
• Deficit justified because net stimulus is real

During expansions (unemployment low, growth strong):

• Run small deficits or surpluses (0-2% of GDP)
• Avoid crowding out
• Private investment drives growth
• Deficit constraint prevents long-term crowding out
• Saves fiscal space for next recession

Long-term debt management:

• Maintain stable or declining debt-to-GDP ratio
• Debt-to-GDP should fall during expansions (surpluses reduce debt)
• Debt-to-GDP will rise during recessions (deficits increase debt)
• Net result over full business cycle: Stable debt-to-GDP ratio

How U.S. Policy Deviates from Optimal

Recent U.S. fiscal policy:

2008-2009 recession:

• Large deficits: 10% of GDP
• Crowding out minimal
• Stimulus appropriate
• Assessment: Correct policy

2010-2019 expansion:

• Deficits averaged 4% of GDP (should have been 0-2%)
• Crowding out moderate
• Stimulus inappropriate during expansion
• Assessment: Suboptimal policy

2020-2021 recovery from pandemic:

• Large deficits: 12-14% of GDP
• Crowding out: Minimal in 2020, increasing in 2021
• Stimulus appropriate in 2020, excessive in 2021
• Assessment: Right policy at wrong time (2021 stimulus too large)

2022-2023 expansion:

• Deficits: 6-7% of GDP
• Crowding out: Strong (full employment)
• Stimulus inappropriate during expansion
• Assessment: Clearly suboptimal policy

Overall assessment: U.S. fiscal policy in 2010s and 2020s has been too stimulative during good economic times and insufficiently counter-cyclical. This suggests current deficits are larger than optimal, creating unnecessary crowding out and long-term growth constraints.

Common Mistake: "Government Borrowing Always Crowds Out Private Investment"

This is too strong. Crowding out varies enormously with economic conditions:

This is correct:

• Crowding out occurs and matters (30-50% in normal times)
• High debt reduces long-term growth
• Large deficits during expansions are problematic

This is incorrect:

• Crowding out occurs in all circumstances (false; minimal in recessions)
• Government borrowing has zero net effect on total investment (false; total investment increases but less than government borrowing)
• Deficits always reduce growth (false; counter-cyclical deficits improve stability)

The nuance:

• Deficits during recessions: Minimal crowding out, stimulus effective
• Deficits during expansions: Strong crowding out, stimulus ineffective
• Sustained high deficits: Long-term crowding out, growth constraint

Real-World Examples: Crowding Out in Different Economies

United States: Global Capital Access Reduces Crowding Out

• Large foreign capital inflows (foreigners buying U.S. Treasuries)
• This supplements domestic loanable funds supply
• Empirical crowding out: 30-40% (lower than pure domestic model would suggest)
• Benefit: Can borrow large amounts without massive interest rate increases

2020-2021 pandemic deficits:

• U.S. borrowed $3+ trillion in 2020-2021
• Interest rates fell (rather than rising) despite massive borrowing
• Reason: Federal Reserve bought Treasury bonds, foreign investors bought Treasuries
• Crowding out: Minimal (near zero in 2020, increasing in 2021)

Japan: 1990s Fiscal Policy Experiment

• Japan ran large deficits (3-8% of GDP) throughout 1990s
• Interest rates remained low (zero bound by mid-1990s)
• Private investment did not recover (continued falling)
• Reason: Deflationary expectations overwhelmed crowding out effect
• Result: Liquidity trap (crowding out occurs, but also zero interest rates and zero stimulus)

Euro Area: Capital Constraints and Crowding Out

• Member countries cannot print their own currency (euro printed by ECB)
• Capital cannot flow freely between countries (imperfect integration)
• Crowding out estimated at 40-50% (higher than U.S.)
• Large fiscal stimulus in one country raises rates across euro area
• Implication: Fiscal policy less effective in euro area than in U.S. or Japan

Related Concepts and Further Reading

• Ch. 8 Stimulus Checks — Government spending affected by crowding out mechanisms
• Ch. 8 War Spending — Historical deficits and crowding out effects
• Ch. 7 Fiscal Policy — Government spending and deficits
• Ch. 9 Monetary Policy — Central bank effects on interest rates and crowding out
• Ch. 10 Growth and Productivity — Capital accumulation and long-term growth
• Federal Reserve economic data (FRED): Interest rates, government borrowing, private investment
• Treasury.gov: Government borrowing and deficit data
• CBO.gov: Congressional Budget Office projections of crowding out effects

Summary: Crowding Out and Its Effects on the Economy

Crowding out is a real and important mechanism by which government borrowing affects private investment. When the government borrows:

In recessions: Crowding out is minimal (0-10%) because loanable funds are abundant. Interest rates may fall. Government borrowing does not displace much private investment. Stimulus is effective.

In expansions: Crowding out is strong (30-50%) because loanable funds are limited and all resources are employed. Interest rates rise substantially. Government borrowing displaces significant private investment. Stimulus is weak or ineffective.

Long-term effect: Sustained high government debt maintains elevated interest rates, reducing private capital accumulation. Lower capital per worker reduces productivity growth. Very high debt (150%+ of GDP) can reduce long-term growth rates by 0.5-1.0 percentage points annually.

Optimal fiscal policy runs large deficits during recessions (stimulus is effective) and small deficits or surpluses during expansions (reduce crowding out). This counter-cyclical approach stabilizes the economy and maintains healthy long-term debt levels.

The U.S. has deviated from optimal policy in recent decades, running elevated deficits during expansions when crowding out is strong and stimulus is ineffective. This suggests current fiscal policy is suboptimal: deficits provide minimal stimulus benefit while creating crowding out that constrains long-term growth.

Next

→ Next chapter: Crypto and digital money primer.