The US government budget deficit will expand substantially in 2009 and 2010. We believe it could reach $1,000bn in 2009 (7% of GDP) and expand further in 2010 (the European Commission forecasts 9% of GDP in 2010, chart 1). Fiscal stabilisers will kick in as the economy enters recession, such that tax revenue will fall and social security payments will increase. The budget deficit for financial year 2008 (which ended in September) exceeded Congressional Budget Office (CBO) forecasts by $48bn, or 12%, due to the weakening economy. CBO forecasts for 2009 appear wildly The Royal Bank of Scotland Group optimistic. Latest releases assume GDP growth of 1.1% and forecast a budget shortfall of $438bn, or 3% of GDP.

We are less optimistic and expect at least a 0.7% contraction in US GDP. As a rule of thumb, each percentage point decline in GDP leads to at least a $40bn increase in the budget deficit. In addition, the increase in the Alternative Minimum Tax (AMT) threshold, recently announced and not yet incorporated into the CBO forecast, would add another $80bn to the shortfall. Before the costs of various financial system bailout packages or additional fiscal stimulus measures are taken into account, we expect a minimum budget deficit in 2009 of $600bn (4.2% of GDP).

Additional fiscal stimulus is likely. Congress can be expected to pass a second stimulus package following the defeat of an earlier package by the Senate in September, and President-elect Obama will follow up with his own policy measures. Mr Obama is expected to introduce a fiscal stimulus to the tune of $200bn to inflate the economy out of recession, in addition to plans to cut the tax burden for working and middle class families and increase spending on affordable healthcare.

The US cannot afford NOT to have the bailout. Without it the financial crisis would be much deeper and longer and the impact on the real economy more devastating.

The Troubled Assets Rescue Plan (TARP) and the Fannie and Freddie rescue are officially off-budget expenditures and will not inflate the budget deficit per se. However, the Treasury will need to raise funds to cover its outlays and this will add to the official government debt. If TARP spending is completely disbursed in the next four quarters as is expected, it will add $700bn to 2009 financing requirements. Excluding the rollover of existing borrowing, the budget deficit, TARP and GSE financing requirements could reach $1½ trillion (10% of GDP).

This exaggerates the potential increase. The proceeds from the eventual sale of mortgage assets and equity in financial institutions would be offset against the initial expenditure. It is perfectly feasible for the Treasury to make a profit out of TARP if asset values rebound. The increase in government debt due to TARP would therefore be a lot lower than $700bn and the overall financing requirement would stay below $1½ trillion.

The experience of Sweden and Japan when they faced similar financial crises is instructive. In 1992 large losses in the banking system following the bursting of a property bubble gave rise to a systemic problem in Sweden. This led the Swedish government to guarantee all bank deposits and creditors of the nation's 114 banks in September 1992. The authorities then took equity stakes in various banks in order to recapitalise them.

Sweden spent around four percent of its GDP to rescue ailing banks. But the final cost to Sweden ended up being less than two percent of GDP. Depending on how certain rates of return are calculated, some estimates suggest it was closer to zero. In particular, financial market confidence in Sweden returned after the government's intervention. The Swedish currency depreciated, but there was no run on the Swedish Krona, and the interest the government needed to pay on its debt (i.e. long-term interest rates) fell (chart 3).

Similarly, the Japanese bailout of its financial system between 1997 and 2002, while ultimately more costly to the Japanese taxpayer than the Swedish experience, did not lead to a significant increase in long-term interest rates. Japan spent nearly $440 billion, mostly from 1998 to 2002, to protect depositors, nationalise the most troubled banks and beef up the capital of other financial institutions. Despite the extra strain on public finances, long-term interest rates remained low.

To fund the budget deficit the US government issues debt. The level of issuance is therefore about to explode. A concern is that there may not be sufficient demand for the debt issuance to be completed in full unless the yield on government debt rises significantly.

Foreign central banks have been willing to fund US treasury acquisition from foreign currency reserves as the US has run a growing current account deficit. This has changed. An appreciating dollar and falling US current account deficit (see our paper on US exports) mean central banks will not accumulate dollar reserves nearly as quickly as they did earlier this decade. The size of the new issuance and reduced level of demand could mean the government will have to offer a higher coupon (interest rate) to entice investors.

This has two consequences:
1. The government's cost of borrowing increases due to the higher interest rate payable on the new debt. Redeemed existing borrowing will need to be rolled over at higher interest rates, further increasing the cost of debt.
2. Companies wishing to raise capital are forced to offer a higher yield than the government as their debt is riskier. The increased cost may make investments unprofitable. Private sector investment could therefore be crowded out by higher public sector borrowing, lowering economic growth in the medium term.

However, it is more likely that in the current environment the government will have no problem finding buyers for its debt. Low interest rates, a poor economic outlook and low risk appetite (reflected in the recent flight to the security of government assets), mean investors are happy to purchase government debt without being compensated with higher yields. The recessionary outlook will be sufficient to dissuade corporations from issuing debt for investment anyway, and low interest rates would make the risk premium over government debt manageable if they did. Consequently private sector investment is unlikely to be crowded out and debt servicing costs may not rise substantially in the short term

In the medium term the picture is bleaker. Problems are likely to emerge in late 2009 to early 2010. When the economy recovers, the Federal Reserve's benchmark interest rate will rise and risk appetite will return to the market. The interest on the stock of government debt will increase and this will place a large burden on public finances. Private investment could be crowded out at this time. An increasing share of income will be devoted to debt repayment, fewer funds will be available for investment, which will harm economic growth.

It is easy to focus on the negatives, but the economy is facing a deep and protracted recession and things would be a lot worse if these measures were not implemented.