What is the crowding-out effect?

The crowding-out effect occurs when government borrowing to finance fiscal spending increases interest rates in the credit market, making it more expensive for private businesses and households to borrow. As a result, private investment in factories, equipment, and real estate declines—not because businesses don't want to invest, but because higher borrowing costs reduce their expected returns. The government's spending boost is partially or fully offset by reduced private investment, shrinking the net stimulus to the economy.

Quick definition: Crowding out is the reduction in private investment caused by higher interest rates resulting from increased government borrowing, which partially or entirely offsets the expansionary effect of fiscal stimulus.

Key takeaways

• Crowding out occurs when government borrowing pushes up interest rates and reduces the cost-effectiveness of private investment
• Partial crowding out means stimulus spending boosts output less than the Keynesian multiplier predicts
• Complete crowding out means fiscal stimulus has zero net effect on output—private investment falls by exactly the amount of government spending
• The strength of crowding out depends on the interest elasticity of investment, monetary policy, and financial market conditions
• Crowding out is weaker when interest rates are already low (near the zero lower bound) and stronger when rates are high
• Empirical evidence suggests partial crowding out is typical, with roughly 20–50% of fiscal stimulus offset by reduced private investment in normal times
• The relationship between government spending and private investment is one of the most contested topics in macroeconomics

How crowding out works: the loanable funds mechanism

To understand crowding out, think of credit markets as a supply-and-demand system. The supply of loanable funds comes from savers (households and institutions). The demand comes from borrowers: private firms wanting to invest in factories and equipment, households seeking mortgages, and governments financing deficits.

In equilibrium, the quantity of loanable funds supplied equals the quantity demanded, and the interest rate clears the market.

Normal times (no fiscal stimulus):

• Private investment demand: high because expected returns exceed borrowing costs
• Government borrowing demand: moderate, financing ordinary operations
• Equilibrium interest rate: r₁

Government enacts large fiscal stimulus through borrowing:

• Government borrowing demand suddenly increases (the government enters the market as a large new borrower)
• Demand for loanable funds shifts outward
• Supply of loanable funds is relatively fixed in the short term (based on savings behavior)
• With demand up and supply flat, the interest rate rises from r₁ to r₂
• At the higher rate r₂, private firms and households find investment less attractive
• Private investment demand falls—the market quantity of private borrowing declines
• The new equilibrium combines higher government borrowing with lower private borrowing

The net effect on total investment is ambiguous: did government borrowing rise by more than private investment fell? If so, fiscal stimulus boosted total demand and output. If private investment fell by more, stimulus was entirely crowded out.

The crowding-out trade-off in debt markets

The mechanism is clearest in long-term debt markets, like the market for corporate bonds and government bonds.

Before stimulus: A 10-year US Treasury bond yields 2%, and a 10-year corporate bond yields 4%. A firm considering a $10 million plant investment expects a 5% annual return. The firm borrows at 4%, earning a 1% spread, and invests.

Government enacts $500 billion stimulus, financed by issuing bonds: Government borrowing surges. Demand for bonds (both government and private) climbs. Bond prices fall, yields rise.

• The 10-year Treasury yield rises to 2.5%
• The 10-year corporate bond yield rises to 4.5%
• The same firm now borrows at 4.5%, expecting a 5% return, earning only a 0.5% spread

The lower spread makes the investment less attractive. The firm might:

• Delay the project (raising expected returns or waiting for rates to fall)
• Scale back the scope (invest $8 million instead of $10 million)
• Abandon it entirely (if the 5% expected return now falls below the 4.5% borrowing cost)

Across many firms, this margin shrinkage translates into significantly lower private investment.

Partial crowding out: the typical case

In most economic conditions, crowding out is partial. Government borrowing increases interest rates and reduces private investment, but not by the full amount of government spending.

Empirical example from the 2008–2009 financial crisis: The US government enacted the American Recovery and Reinvestment Act (ARRA), a $787 billion fiscal stimulus. The Keynesian multiplier might have predicted an output boost of $787 billion × 1.2 = $944 billion. However, if 40% crowding out occurred:

• Gross Keynesian effect: +$944 billion
• Crowding out (private investment reduction): −$315 billion (40% of $787 billion)
• Net effect: $944 billion − $315 billion = $629 billion in output growth

The government spent $787 billion but net output only grew by $629 billion, implying a net multiplier of 0.8 instead of 1.2. Crowding out reduced the effective stimulus by one-third.

Complete crowding out: the Ricardian case

Some economists, notably Robert Barro, argue that fiscal stimulus might cause complete crowding out, meaning private investment falls by exactly the amount of government spending, leaving total demand unchanged.

This occurs under two conditions:

1. Perfect interest-rate response: Government borrowing is so large relative to the credit market that it drives up interest rates substantially. Private firms find borrowing prohibitively expensive and cancel most investment projects.

2. Forward-looking households (Ricardian equivalence): Households recognize that larger government deficits today mean higher taxes or inflation tomorrow. They increase saving to compensate, reducing consumption and demand. Higher savings increase the loanable-funds supply, preventing interest rates from rising as much, but also reducing the multiplier because households cut consumption.

Under Ricardian equivalence, households offset the stimulus through higher saving, and private investment falls due to the forward-looking tax burden. The net effect is zero: government spending up by $100 billion, private consumption and investment down by a combined $100 billion.

Evidence on Ricardian equivalence is mixed. Households in practice do not perfectly foresee future tax burdens or do not adjust saving perfectly. But some partial Ricardian behavior might occur, especially in high-credibility environments where households trust the government's long-term fiscal sustainability. This could explain why multipliers vary across countries and time periods.

When is crowding out weak? Interest rates near zero.

Crowding out is weakest when interest rates are already near zero because further borrowing by the government cannot push rates lower. If the Federal Reserve has already set short-term rates to near zero, long-term rates are still positive but constrained by the zero lower bound on short rates. Government borrowing has little room to drive rates higher.

The 2008–2009 financial crisis example:

• Federal Reserve cut the federal funds rate to near 0% in late 2008.
• Even with massive government borrowing, 10-year Treasury yields stayed below 3%.
• Private firms still found borrowing cheap and continued some investment.
• Crowding out was minimal because interest rates couldn't rise much.
• The Keynesian multiplier was closer to its theoretical value.

The 2020 COVID-19 pandemic:

• Federal Reserve cut rates to zero and began quantitative easing.
• The Treasury issued over $3 trillion in debt in 2020–2021 to fund stimulus.
• Long-term rates initially fell (because of Fed purchases), then gradually rose as recovery expectations improved.
• Crowding out was again muted in the immediate aftermath because the Fed kept rates low and even purchased Treasuries to suppress yields.

This illustrates a key point: monetary policy and fiscal policy interact. The central bank can mitigate crowding out by keeping rates low, enabling both government and private borrowing. Conversely, if the central bank tightens while the government borrows heavily, crowding out intensifies.

When is crowding out strong? High interest rates and competitive borrowing.

Crowding out is strongest when interest rates are already high and the credit market is tight.

Historical example: 1980s government borrowing in the US:

• The US federal deficit surged to 5–6% of GDP under Ronald Reagan's supply-side tax cuts and military spending.
• The Federal Reserve under Paul Volcker was tightening to fight inflation.
• Interest rates rose sharply: 10-year Treasury yields reached 13–15%.
• Private investment plummeted. Business capital investment as a share of GDP fell from 10% (1970s) to 7% (1982–1985).
• The strong private-investment crowding out, combined with high inflation expectations, meant the fiscal stimulus had a muted effect on real output.

Why crowding out was strong:

• No monetary accommodation: the Fed was fighting inflation, not supporting stimulus.
• High baseline rates: interest rates had nowhere to go but further up.
• Elastic investment demand: firms were highly sensitive to borrowing costs, so they quickly canceled projects.

In this scenario, an extra $100 billion of government spending might have increased output by only $40–50 billion, a multiplier well below one.

The transmission through financial markets

Crowding out operates through multiple channels:

1. Corporate bond markets Government bond issuance increases, borrowing costs for corporations rise, and companies reduce capital expenditures and hiring.

2. Bank lending Banks that hold government bonds see their portfolio yields rise, but they also face competition from government securities. They may tighten lending standards on risky private loans, reducing private credit availability. This is a financial accelerator mechanism: tight private credit reinforces the impact of higher rates.

3. Stock markets Higher interest rates reduce the present value of future corporate profits. Stock valuations fall. Firms raise capital more expensively through equity issuance and reduce investment. This is often called the "wealth effect" or Tobin's Q (the ratio of asset values to replacement costs).

4. Exchange rates (open economies) Higher interest rates attract foreign capital inflows. Demand for the domestic currency rises, appreciating the exchange rate. Exports become less competitive, reducing export demand and manufacturing investment. This external crowding out is particularly important for small, open economies like Canada or Australia.

Empirical evidence on crowding out

The magnitude of crowding out remains contested among economists.

Estimates from major studies:

Study / Period	Estimated Crowding Out	Interpretation
Barro (1989), long-run	100%	Fiscal stimulus has no permanent effect; Ricardian equivalence
Auerbach & Gorodnichenko (2012), US recession stimulus	20–30%	Partial crowding out during downturns
IMF (2010), cross-country analysis	25–50%	Typical crowding out in normal times; less in recessions
Ramey (2011), US military spending shocks	30–60%	Defense spending crowds out private investment partially
Bivens (2015), ARRA (2009)	15–40%	Low crowding out due to low interest rates and slack
Cutting et al. (2020), COVID stimulus	10–30%	Very low crowding out due to zero rates and QE

Why estimates vary:

• Different time periods have different interest-rate environments and expectations.
• Different countries have different financial-market structures (bank-based vs. capital-market-based credit systems).
• Different fiscal programs have different compositions (transfers, public investment, tax cuts) and different crowding-out potentials.
• Methodological differences in identifying causal effects (simulations vs. natural experiments vs. regression analysis).

Consensus view: In normal times with moderate interest rates, crowding out is partial, offsetting 20–50% of fiscal stimulus. In recessions with low rates, crowding out is weak, offsetting less than 20%. In booms with high rates, crowding out is strong, offsetting 50%+ of stimulus.

Crowding out in different economic sectors

Crowding out doesn't affect all private investment equally.

Sectors most vulnerable to crowding out:

• Manufacturing and capital-intensive industries depend heavily on long-term borrowing. Higher rates immediately reduce expected returns.
• Construction and real estate are extremely interest-rate-sensitive. A 1% rise in mortgage rates can cut housing investment by 10–20%.
• Small businesses lack access to equity markets and rely on bank loans. Tighter lending standards during credit crunches hit them hard.

Sectors less vulnerable to crowding out:

• Technology and high-growth companies often operate on retained earnings or venture capital. They are less sensitive to bond-market interest rates.
• Utilities and regulated infrastructure have stable, predictable cash flows. They can invest even at higher rates because returns are guaranteed by regulation.
• Intangible investment (R&D, software) requires less external financing than physical capital.

Government stimulus that crowds out construction and manufacturing is economically costly because these sectors have large multiplier effects and create jobs for workers without college degrees. Crowding out of high-growth tech investment is less damaging to near-term employment.

The investment-savings identity and crowding out

From a national accounting perspective, total savings must equal total investment (including government deficits). If government deficits rise and national savings don't change, private investment must fall by an equivalent amount—this is an identity, not a theory.

Accounting identity:

National Savings = Private Savings + Government Savings
Total Investment = Private Investment + Government Investment (or deficit)

If: National Savings = constant
And: Government deficit increases (negative government savings)
Then: Private Investment must decrease

This identity shows that some crowding out is mechanical—it must occur in national accounts. The question is whether the identity is exact (100% crowding out) or whether other channels (foreign capital inflows, changes in savings behavior) allow some room for net fiscal expansion.

In open economies with capital flows: If a country runs a fiscal deficit, foreign investors might purchase government bonds, bringing in capital. The capital inflow finances both government and private investment, reducing but not eliminating crowding out. The US benefits from this: large fiscal deficits are partially financed by foreign capital inflows, which reduces crowding out of private investment.

How central banks can mitigate crowding out

The Federal Reserve and other central banks can reduce crowding out through:

1. Monetary accommodation Keep short-term interest rates low or negative (nominal or real), limiting upward pressure on long-term rates. This was the strategy in 2008–2009 and 2020–2021.

2. Quantitative easing (QE) Directly purchase government and private bonds, suppressing yields and expanding credit availability. This also mitigates crowding out by lowering long-term rates despite high government issuance.

3. Forward guidance Signal that interest rates will remain low for an extended period, anchoring long-term rate expectations and preventing crowding out through the expectations channel.

4. Negative interest rates Central banks in some countries (European Central Bank, Bank of Japan) have imposed negative rates on reserve deposits, making holding cash expensive and encouraging bank lending rather than government-bond hoarding.

During the COVID-19 pandemic, for example:

• Federal Reserve cut rates to zero.
• Fed launched massive QE, purchasing $700+ billion in bonds in weeks.
• Treasury yields fell despite massive government borrowing.
• Private credit markets remained liquid, and private investment wasn't heavily crowded out.

This demonstrates that crowding out can be mitigated or eliminated if the central bank is willing to accommodate fiscal stimulus.

Real-world examples of crowding out

1. US fiscal expansion, 1983–1989 Reagan tax cuts and defense spending surged the deficit to 5% of GDP. Volcker's Fed maintained tight monetary policy to fight inflation. Interest rates rose sharply, 10-year Treasuries exceeded 13%, and private business investment fell. Crowding out was substantial, and real output growth was moderate (3–4% annually) despite large deficits.

2. Japan's Lost Decade, 1990s The Japanese government ran large deficits to stimulate the economy after the real estate bust. Despite zero short-term interest rates, long-term interest rates stayed sticky around 1–2%. Private investment didn't rise because businesses were deleveraging (paying down debt) rather than investing. Crowding out was partial; the spending didn't translate to output growth because of Ricardian behavior and weak private demand.

3. Eurozone austerity, 2010–2015 In response to high deficits, several European governments cut spending sharply. With the European Central Bank keeping rates positive and tight, interest rates didn't fall much despite lower government borrowing. Private investment also fell due to weak demand and credit-supply constraints. The austerity crowded itself out; cutting government spending didn't free up resources for private investment because private demand was weak.

4. UK post-GFC recovery, 2009–2015 The Bank of England maintained near-zero rates and quantitative easing despite high government borrowing. Long-term interest rates stayed low (2–3%), and private investment recovered gradually. Crowding out was minimal because monetary policy was accommodative. The recovery was steady but slow, reflecting supply-side damage rather than demand-side crowding out.

Common mistakes

1. Assuming all crowding out is bad. Some crowding out is inevitable and even desirable. If government spending is on productive public investment (schools, roads) and it crowds out unproductive private investment (speculative real estate), the net effect might be positive for long-term growth.

2. Forgetting that crowding out varies with interest rates. Claiming "fiscal stimulus always crowds out private investment" ignores that crowding out is weak at zero interest rates. The timing and monetary policy context matter enormously.

3. Confusing crowding out with Ricardian equivalence. Ricardian equivalence is a specific mechanism (households anticipate future taxes and save more). Crowding out is more general (private investment falls due to higher interest rates). They can occur together, but they are not the same.

4. Using long-run studies to criticize short-run stimulus. Long-run crowding out can be substantial (perhaps 80–100%) if private investment eventually adapts to permanently higher interest rates. But in the short run, when most unemployment exists, crowding out is weak, and stimulus is effective. Mixing time horizons confuses the debate.

5. Ignoring differences in investment composition. Stimulus that crowds out speculative real estate or consumption-smoothing borrowing has less negative impact than stimulus that crowds out productive equipment investment or R&D. The quality of what's crowded out matters.

External resources

• Interest Rates and Fiscal Policy – Federal Reserve: The Fed's research on how government borrowing affects interest rates and credit market dynamics.
• Bond Yields and Economic Data – FRED: The Federal Reserve Economic Data service provides historical Treasury yields and data for analyzing crowding-out mechanisms.

FAQ

Can crowding out be negative (crowding in)?

Yes. In rare cases, government investment can boost private investment through crowding in. For example, government spending on infrastructure (roads, ports, electricity) lowers private firms' transportation and logistics costs. This can raise expected returns on private investment and increase private borrowing and investment despite higher interest rates. This is explored in the next article.

Does crowding out happen immediately or gradually?

Crowding out happens gradually. Interest rates rise immediately, but firms typically have long lead times for investment decisions (planning, permitting, construction). Private investment decisions adjust over 6–18 months. Statistical crowding out is most evident 1–2 years after fiscal stimulus begins.

If the central bank targets inflation, not interest rates, does crowding out change?

Yes, in principle. An inflation-targeting central bank might tolerate higher interest rates if stimulus inflation expectations rise. This could increase crowding out. However, most modern central banks use interest rates as their operational target, so the practical difference is small.

What's the relationship between crowding out and government debt sustainability?

Crowding out reduces the immediate stimulus from fiscal spending, but it doesn't directly affect long-run debt sustainability. However, if crowding out reduces private investment and long-run productivity growth, the tax base grows slower, making high debt less sustainable. Crowding out has long-run fiscal consequences.

Can fiscal stimulus and monetary policy both be tight without causing severe crowding out?

This is contradictory. If monetary policy is tight (high interest rates), fiscal stimulus faces strong crowding out. If fiscal policy is tight (low spending), there's nothing to crowd out. Effective stimulus requires at least one of the two policies to be expansionary.

Related concepts

• Learn about the inverse relationship: crowding in (when government investment boosts private investment): ../chapter-08-fiscal-policy/09-crowding-in-effect
• Understand the Keynesian multiplier and how crowding out reduces its effectiveness: ../chapter-08-fiscal-policy/07-keynesian-multiplier-math
• Explore how monetary policy and interest rates drive crowding out: ../chapter-07-monetary-policy/02-interest-rates-inflation
• Examine the relationship between government spending and private borrowing in budget deficits: ../chapter-08-fiscal-policy/10-budget-deficit-explained
• Learn how inflation expectations amplify crowding out through bond markets: ../chapter-04-inflation-deep-dive/01-what-is-inflation
• Discover how international capital flows reduce crowding out in open economies: ../chapter-09-international-trade/01-international-trade-basics

Summary

The crowding-out effect demonstrates that fiscal stimulus's impact on output depends critically on credit-market conditions and monetary policy. When government borrowing raises interest rates significantly, private investment falls, offsetting some or all of the Keynesian multiplier effect. In normal times, partial crowding out (20–50%) is typical. At zero interest rates with monetary accommodation, crowding out is minimal and stimulus is highly effective. The strength of crowding out varies by sector, country, and time period, making empirical estimates highly context-dependent. Recognizing crowding out does not invalidate fiscal policy—it highlights that stimulus works best when combined with monetary accommodation and when interest rates are low.

Next

→ The crowding-in effect