Something bad is happening in the world’s bond markets. Governments are finding it increasingly expensive to borrow money, at precisely the time when they are borrowing enormous amounts of it. In Britain, yields on long-dated government bonds have climbed to levels not seen for decades, while in the United States the federal government faces a growing bill for servicing its outstanding debt. This isn’t simply a British problem, whatever impression you might get from reading some of our newspapers. Government borrowing costs have been rising across much of the developed world. The UK is bringing down is budget deficit yet still finds yields rising. It is not alone:
The immediate consequence is straightforward; higher borrowing costs mean governments must spend more money paying interest on their debts, leaving less available for everything else. 7% of UK Government spending is debt interest. In the US it is closer to 15%
Some economists have long argued that government borrowing creates another problem, potentially even more damaging to economic growth. By competing with businesses for finance, governments supposedly reduce the amount of money available for private investment. This is known as crowding out, and it is one of the principal arguments advanced by those who believe governments should borrow less.
I’ve always been rather sceptical about this argument. It seems to assume that businesses and governments are queuing up at the same counter, competing for a limited supply of money. Anyone who has run a business knows that raising finance is rather more complicated than that. Most businesses don’t sell bonds to international pension funds, and very few have anything to do with the financial markets in which governments raise their money. So how much private investment is really being crowded out, and could government borrowing sometimes have precisely the opposite effect?
The rising cost of government borrowing
Governments borrow money by issuing bonds, known in Britain as gilts and in America as Treasury securities. Investors lend money to the government in return for interest payments and the repayment of their original investment when the bond matures. These bonds can subsequently be bought and sold, and their market prices fluctuate according to the returns investors are prepared to accept. When bond prices fall, their yields rise, which means that governments must generally offer higher interest rates when they issue new debt.
The distinction between the price of a bond and its yield is important. If the government issued a conventional fixed-rate gilt ten years ago, the interest payment on that bond doesn’t change simply because market yields have increased. But governments are constantly borrowing new money and refinancing existing debts as they mature, so higher yields gradually feed through into their interest bills. Britain also has a substantial stock of inflation-linked government debt, which creates an additional exposure to changes in inflation.
Consider a government borrowing £100 billion. At an interest rate of 3%, the annual interest bill would be £3 billion. Increase the rate to 5%, and that bill becomes £5 billion, an additional £2 billion every year without the government receiving anything more for its money. Multiply that across the hundreds of billions of pounds that governments borrow and refinance, and the consequences for public finances become substantial.
Britain’s outstanding public debt is approaching £3 trillion, while American federal debt held by the public exceeds $30 trillion. Both countries are consequently sensitive to changes in borrowing costs, although the effects emerge over time rather than immediately. In America, the Congressional Budget Office projected that federal interest payments would approach $1 trillion in 2026, with further substantial increases expected over the following decade. Interest is becoming one of the largest items in the federal budget, despite providing no new public services or infrastructure.
Why are investors demanding higher returns?
Part of the explanation is inflation, which reduces the real value of the money investors receive when their bonds mature. Central banks have also been reducing the enormous holdings of government bonds they accumulated through quantitative easing, leaving private investors to absorb more of the debt being issued. Meanwhile, governments continue to run substantial deficits, increasing the supply of bonds coming onto the market. Investors have become less willing to lend money for long periods without receiving additional compensation for the uncertainty involved.
The international nature of these markets is particularly important. American Treasury securities provide a benchmark for interest rates around the world, so developments in the United States can influence borrowing costs in Britain even when nothing particularly significant has changed in the British economy. It would be a mistake to interpret every increase in gilt yields as a judgement on the Chancellor’s economic policies.
Higher interest payments are an obvious problem for governments, but there is another potential consequence. If governments are borrowing more money and offering higher returns to attract investors, what happens to businesses that also need finance?
Does government borrowing crowd out private investment?
The traditional economic explanation is that there is a limited supply of savings available for investment. Governments and businesses compete for those savings, and when governments borrow more, the increased demand pushes up interest rates. Some businesses consequently find that investments which would previously have been profitable are no longer worthwhile. Factories aren’t built, machinery isn’t purchased and businesses expand more slowly than they otherwise would.
This is the conventional explanation of crowding out, and it sounds perfectly reasonable until you start looking at how businesses actually finance themselves. The argument assumes that government borrowing and business investment are drawing on substantially the same sources of money, but that isn’t necessarily the case.
Consider a small manufacturing business wanting to purchase a new machine. It might use its accumulated profits, arrange a bank loan or lease the equipment from a specialist finance company. A restaurant opening another branch might rely on the owner’s savings and a commercial mortgage, while a technology start-up might seek investment from venture capitalists or wealthy individuals. Growing businesses have access to private equity funds, family offices and crowdfunding platforms, while larger companies can raise money by issuing new shares or selling corporate bonds to institutional investors.
These are quite different sources of finance, involving investors with different objectives and appetites for risk. Someone providing venture capital to an early-stage technology business is hoping to make a substantial return on a risky investment. Someone purchasing a thirty-year government bond is generally looking for a relatively predictable stream of income. It isn’t immediately obvious that these investors are competing for the same opportunities.
Indeed, many businesses don’t need external finance at all. The Bank of England surveyed nearly 2,900 British small and medium-sized businesses about their investment decisions and found that more than 80% of those undertaking investment had used at least some internal funds. Half had financed their investment entirely from their own resources, while many indicated that they would prefer slower growth to taking on additional borrowing.
That finding is important, although it needs to be interpreted carefully. It doesn’t mean that half of all British business investment is financed from retained profits, because the survey measured the number of businesses rather than the value of their investments. A relatively small number of very large companies can account for a substantial proportion of national investment expenditure. Nevertheless, it demonstrates that many businesses finance investment without going anywhere near the financial markets in which governments borrow money.
If a manufacturer uses £100,000 of accumulated profits to purchase new machinery, the government hasn’t directly deprived that business of the money it needs. The same applies to a restaurant owner investing personal savings or a small business raising money from friends and family. Whatever the wider consequences of government borrowing, these businesses aren’t competing directly with the Treasury for finance.
To establish where competition really exists, we need to examine who actually purchases government bonds.
According to HM Treasury, overseas investors held approximately a third of outstanding British government bonds in September 2025. Insurance companies and pension funds owned around 21%, while the Bank of England’s Asset Purchase Facility held approximately 19%. The remaining 27% belonged to banks, investment funds and other investors.
There is clearly some overlap between these investors and the institutions financing British businesses. Pension funds and insurance companies purchase corporate bonds as well as government securities, while investment funds hold shares in companies alongside their holdings of gilts. Banks also hold government bonds while lending money to businesses, and some hedge funds participate in both markets.
But the overlap is far from complete. A foreign central bank holding British government bonds as part of its international reserves isn’t considering an alternative investment in a manufacturing business in County Durham. A pension fund purchasing long-dated gilts to match its future pension liabilities isn’t choosing between those bonds and financing a new technology company. Nor is someone contributing £100 to a crowdfunding campaign lweighing up the return available on government debt. This is the real world.
There is some competition in the corporate bond market, where large businesses borrow directly from institutional investors. A pension fund might choose between purchasing a government bond yielding 5% or a corporate bond yielding 6%. The company must offer a higher return because lending to it involves greater risk. If government bond yields increase to 6%, the company will generally have to offer a higher rate to persuade investors to buy its bonds. Here the crowding-out mechanism is quite direct, because governments and companies are competing for investment from many of the same institutions.
Something similar can happen in equity markets. Institutional investors hold both government bonds and company shares, and higher bond yields can make equities less attractive by comparison. But even here the relationship is less straightforward than it first appears. Most transactions on the stock market involve investors buying existing shares from one another, rather than providing new money to businesses. Only new share issues directly raise additional equity finance, although changes in share prices can influence the cost of doing so.
Private equity and venture capital are further removed from the government bond market, although their underlying investors may include pension funds and other institutions that also purchase gilts. The relationship becomes more distant still when we consider crowdfunding, owners’ savings and retained profits. Even where the same ultimate investors are involved, the decisions they are making may be quite different.
Crowding out is principally a problem for a very small number of larger businesses raising money through corporate bonds and institutional capital markets. For smaller businesses, which rely much more heavily on retained profits and bank lending, the connection doesn’t exist. Business schools and economics departments teach crowding out because they are influenced by, and some times funded by, the small number of very big companies for whom crowding out is an issue. They miss the vast majority of businesses for whom crowding out doesn’t exist.
But before dismissing the argument altogether, we need to look more carefully at the banks.
What about fractional reserve banking?
The traditional explanation of banking is that customers deposit money, banks retain a proportion as reserves and lend the remainder to borrowers. The money that is lent eventually finds its way back into the banking system as further deposits, allowing more lending to take place. This is the familiar account of fractional reserve banking, and it suggests that banks are essentially intermediaries between savers and borrowers. The reserve that they hold consists of cash and liquid assets such as Government bonds. Banks hold government bonds because they are relatively safe and liquid assets, which can be sold or used as collateral when the bank needs funding. A reduction in government bond issuance reduces the availability of an asset that helps banks manage their liquidity, rather than simply releasing additional money for business lending. Fewer bonds = less lending.
Given that a lot more business borrow from banks than from bond markets reducing the amount of Government bonds issued reduces the amount available for investment. The real crowding out comes from the housing market. Banks make choices about how much to lend to consumers (personal loans and credit cards), how much to loan to businesses and how much to devote to mortgages. A continual house price boom diverts money from business to houses.
But in 2014 the Bank of England published a paper entitled Money Creation in the Modern Economy, which challenged the conventional explanation of how commercial banks operate. The Bank argued that banks do not simply lend out money deposited by savers. Instead, when a bank approves a loan, it normally creates a corresponding deposit in the borrower’s account. Lending creates deposits, rather than deposits mechanically determining the amount that can be lent.
This is a rather fundamental distinction. A bank doesn’t need to wait for someone to deposit £100 before it can lend £90 to a business. It can create a new loan and a corresponding deposit, subject to the constraints imposed by its capital, liquidity, funding costs and the creditworthiness of its borrowers. The familiar idea of a fixed pool of deposits being multiplied through the banking system is therefore a misleading description of how modern banking works.
At first sight, this appears to strengthen the case against crowding out. If banks create money when they lend, why should additional government borrowing necessarily reduce the amount available to businesses? And if a business finances its investment from a newly created bank loan, rather than from money previously saved by someone else, where exactly is the competition with the government?
But the Bank of England’s explanation challenges my original scepticism. Banks may create deposits when they lend, but they cannot create unlimited amounts of credit without regard to the cost of funding, their regulatory requirements or the returns available elsewhere in financial markets. Government bond yields influence those calculations, even where the bank is not directly choosing between buying a gilt and making a particular business loan.
Government bonds provide a benchmark against which many other financial assets are priced. If investors can earn a higher return from lending to the government, they will generally demand higher returns from riskier investments. Banks face changing funding costs and market interest rates, while the value of the securities they hold also fluctuates. These effects can feed through into the rates charged to businesses, including those that have never issued a bond or raised money from institutional investors.
The same principle applies beyond banking. Venture capitalists and private equity investors may not regard government bonds as direct alternatives to the businesses they finance, but the returns available on relatively safe investments still influence the valuations they are prepared to accept for risk. Higher interest rates also increase the cost of debt used in many private equity transactions, potentially reducing the number of investments that are commercially attractive.
The Bank of England’s analysis therefore undermines the simplest explanation of crowding out without eliminating crowding out itself. Governments and businesses aren’t necessarily competing for a fixed quantity of savings, but government borrowing influences the price of capital throughout the economy. A small business may never compete directly with the Treasury for investment, yet still face higher borrowing costs because government bond yields have increased.
That doesn’t establish how much investment is actually displaced. We don’t have a reliable breakdown of all British business investment according to its ultimate financing source, and higher government bond yields can reflect inflation expectations, monetary policy and economic growth rather than simply the amount governments borrow. It would therefore be difficult to calculate how much private investment has been lost specifically because of government borrowing.
Nevertheless, I think the evidence means that while crowding out is a very outdated economic theory it shouldn’t be completely dismissed. The mechanism is more complicated than the textbook explanation suggests, and its effects are likely to vary considerably between businesses, but the connection between government borrowing costs and private investment is real.
There is, however, another side to the argument. Government borrowing doesn’t simply influence the cost of private investment. Depending on what the money is spent on, it can also make private investment more attractive.
Can government borrowing encourage private investment?
Economists call this crowding in, and the principle is straightforward enough. Businesses don’t make investment decisions solely on the basis of how much it costs to borrow money. They also consider whether the investment will be profitable, which depends on everything from the availability of skilled workers to the quality of transport infrastructure and the reliability of electricity supplies.
Imagine a company contemplating building a new factory. Cheap finance is of little use if the factory cannot obtain an adequate electricity connection, its workers cannot travel to the site or the local infrastructure is incapable of supporting the development. Government investment that removes these obstacles makes projects commercially viable, encouraging businesses to invest money that would otherwise have remained unspent.
The same principle applies to investment in education, research, transport and communications. Government spending increases the productive capacity of the economy, improves the profitability of private investment and creates opportunities that businesses would not otherwise have pursued. In these circumstances, the government is not simply competing with businesses for capital, it creates the conditions in which businesses are willing to invest.
There is evidence that this happens. Research examining 34 OECD countries between 1995 and 2019 found that public investment was associated with increased private investment over the medium to long term, with the effects particularly strong in areas such as economic infrastructure and education. The estimates suggested that additional private investment could eventually exceed the original public investment, although the results varied considerably between countries and types of expenditure.
This doesn’t mean that all government investment is productive, or that every pound borrowed by the government generates additional private investment. A badly designed infrastructure project can consume resources without producing sufficient economic benefits, while public investment undertaken when the economy is already operating at full capacity may simply compete with private businesses for workers, equipment and materials. The important distinction is between spending that increases the economy’s productive capacity and spending that merely increases demand for resources that are already scarce.
Britain’s current investment programmes provide an interesting opportunity to examine the difference. The government’s plans for cleaner electricity require substantial investment in generating capacity, transmission networks, storage and associated infrastructure. Its Clean Power 2030 programme envisages investment of approximately £40 billion annually between 2025 and 2030, with much of that money expected to come from the private sector. The government’s role is not simply to build and operate everything itself, but to create the conditions in which private investors are prepared to commit capital.
This is particularly important in energy, where projects frequently involve enormous initial expenditure, lengthy development periods and uncertainty about future revenues. A wind farm developer needs confidence that electricity can be transmitted to customers and that the project will generate sufficient income over its operating life. A manufacturer considering building a factory to supply equipment for renewable energy projects needs confidence that there will be sufficient demand to justify the investment. Government policy, infrastructure investment and financial guarantees can influence both decisions.
Great British Energy and the National Wealth Fund are intended to help address these problems by investing alongside private institutions and reducing some of the risks that might otherwise discourage investment. The National Wealth Fund reported in October 2025 that its portfolio of £7.5 billion in investment commitments had mobilised £16.2 billion of private investment. That represents more than £2 of associated private investment for every £1 committed by the Fund.
We need to be careful about what these figures demonstrate. Private investment taking place alongside public investment isn’t necessarily investment that would never have happened without government intervention. Some projects might have proceeded anyway, while others might simply have obtained finance from different sources. Nevertheless, the figures demonstrate that public and private investment can be complementary rather than competing activities.
There is an even more interesting example involving smaller businesses. Earlier this month, the British Business Bank published an evaluation of its Nations and Regions Investment Funds, which had invested £294 million of public money alongside £233 million of private investment, supporting 877 businesses. More importantly, 81% of surveyed recipient businesses reported that they would either have been unable to obtain finance without the programmes, or would have obtained less finance or experienced delays.
That is particularly significant because these are precisely the sorts of businesses that frequently struggle to obtain external investment. Rather than displacing private finance, public investment appears to have helped overcome some of the barriers preventing businesses from raising the money they needed. The private investment accompanying the public funding cannot all be assumed to be additional, but the evidence suggests that government intervention has enabled at least some investment that would otherwise have been delayed or abandoned.
Of course, the government still needs to finance its contribution, and additional borrowing may have consequences for interest rates elsewhere in the economy. But if £1 of government investment helps attract additional private capital and generates productive assets that would not otherwise have existed, the economic consequences are rather different from those implied by the conventional crowding-out argument.
America: government borrowing and the technology boom
The United States provides an interesting contrast, because its government is borrowing enormous amounts of money at the same time as American technology companies are undertaking one of the largest private investment programmes in modern economic history.
The scale of investment in artificial intelligence is extraordinary. Major technology companies are spending vast sums on data centres, computing equipment, semiconductor manufacturing and the electricity infrastructure needed to support them. Much of this investment is financed from their own substantial profits, although corporate borrowing is becoming increasingly important as the scale of expenditure grows.
At the same time, the federal government continues to run substantial budget deficits. The Congressional Budget Office projected borrowing of approximately $1.9 trillion in 2026, equivalent to 5.8% of GDP, with federal debt held by the public exceeding 100% of national income. The cost of servicing that debt is also increasing, creating further pressure on future budgets. If the US economy is growing by 2.2% in the middle of an AI boom, while the Government is spending 7% more than it brings in then something is going badly wrong.
President Trump’s tax and spending policies have contributed to the negative fiscal outlook, although it would be misleading to attribute the entire deficit to his administration. America entered his second presidency with substantial structural deficits, accumulated debt and growing interest costs. Nevertheless, the administration’s willingness to continue borrowing heavily raises questions about the consequences for private investment at a time when American businesses are demanding enormous amounts of capital.
The potential competition is particularly obvious in corporate bond markets. Large technology companies raise finance from many of the same pension funds, insurance companies and investment institutions that purchase American Treasury securities. If the federal government must offer higher yields to attract investors, companies issuing bonds may also have to offer higher returns. In principle, the government’s borrowing could therefore increase the cost of financing America’s technological expansion.
But the evidence is not as straightforward as this explanation suggests. Research published by investment manager PIMCO in September examined whether large debt issues by technology companies were pushing up Treasury yields. Its researchers found little systematic evidence that individual corporate bond announcements caused significant increases in government borrowing costs. The results don’t rule out longer-term competition for capital, but they suggest that the relationship between government and corporate borrowing is more complicated than a simple contest for a fixed supply of money.
There is also the question of what the borrowing is being used to finance. American technology companies are investing in new productive assets, although whether all those investments will ultimately prove profitable remains to be seen. A substantial proportion of federal borrowing, by contrast, finances the gap between current government expenditure and taxation, including the growing cost of servicing existing debt. Biden spent $1.2 trillion on the Infrastructure Investment and Jobs Act (2021), $369bn in initial climate and energy provisions through the Inflation Reduction Act (2022), and $52.7bn in semiconductor incentives through the CHIPS and Science Act (2022). All of these schemes grow US infrastructure. Trump’s tax cuts have cost $4.6–5.2 trillion, all funded through debt.
This distinction matters because borrowing to finance current expenditure doesn’t necessarily generate the additional productive capacity that might encourage private investment. That isn’t to suggest that government spending on healthcare, education or other public services has no economic value, or that all private investment is productive. But the economic consequences of borrowing depend partly on whether the expenditure creates additional capacity, removes barriers to growth or simply sustains existing commitments.
America’s experience also illustrates why we need to be cautious about interpreting rising bond yields. The enormous investment programmes undertaken by technology companies may themselves be contributing to higher demand for capital, independently of government borrowing. Stronger expectations for future economic growth can also increase real interest rates, while inflation and monetary policy influence the returns investors require. Higher Treasury yields therefore cannot automatically be attributed to Trump’s fiscal policies, even if those policies are contributing to the pressure.
The interesting contrast with Britain is not that one government crowds in investment while the other crowds it out. Both countries have productive and less productive forms of public expenditure, and both face the consequences of rising borrowing costs. But Britain is explicitly using public financial institutions and infrastructure programmes to encourage additional private investment, particularly in energy and smaller businesses. America, meanwhile, is running very large fiscal deficits alongside an extraordinary private-sector investment boom, raising questions about whether government borrowing is adding to the cost of financing that expansion.
So does government borrowing crowd out economic growth?
I started by questioning whether government borrowing really deprives businesses of the money they need to invest. Having looked at the evidence, I think that scepticism was partly justified, but the answer is more complicated than I initially expected.
The conventional explanation of crowding out, in which governments and businesses compete for a fixed pool of savings, doesn’t adequately describe the way modern businesses finance investment. Many smaller businesses use retained profits or their owners’ resources, while banks create credit rather than simply lending out existing deposits. Direct competition with government borrowing is clearest among larger companies raising money through institutional capital markets, although even there the relationship is not always straightforward.
But the financial system is interconnected, and government bond yields influence the cost of capital well beyond the institutions that actually purchase government securities. A small business borrowing from its bank may face higher interest rates even though it has never competed directly with the government for finance. The Bank of England’s explanation of modern banking therefore complicates the traditional argument without making crowding out disappear.
What the conventional argument frequently overlooks, however, is that government investment can also encourage private investment. Public spending on electricity networks, transport, education and other productive infrastructure can make projects commercially viable that would otherwise never proceed. Britain’s public investment programmes provide evidence that government finance can attract private capital, particularly where businesses face genuine barriers to obtaining investment.
This doesn’t mean that governments can borrow without limit, or that every infrastructure project will generate sufficient economic growth to justify its cost. Higher interest payments are a real burden on public finances, and badly directed government spending can consume resources that might have been used more productively elsewhere. But neither does it follow that every pound borrowed by the government is a pound unavailable to private enterprise.
Borrowing £10 billion to finance existing expenditure is economically different from borrowing £10 billion to build infrastructure that encourages businesses to invest another £20 billion. Both increase government debt, and both may influence borrowing costs, but their consequences for future economic growth could be very different.
Perhaps the debate about government borrowing has been focused too narrowly on the amount of debt governments accumulate. The cost of servicing that debt matters, particularly when bond yields are rising, but so does what governments actually do with the money. If borrowing simply increases competition for scarce resources, crowding out is a genuine risk. If it removes obstacles to private investment and increases the productive capacity of the economy, the result may be precisely the opposite.
The important question isn’t simply how much governments borrow, but whether that borrowing makes it easier or harder for businesses to invest, expand and grow.
https://www.bankofengland.co.uk/quarterly-bulletin/2014/q1/money-creation-in-the-modern-economy