The federal fiscal deficit fell sharply in Biden’s first year and a half in office.  This was largely due to the expiration of the huge Covid relief programs that had been approved both during Trump’s last year in office and then at the start of Biden’s term (totaling $5.7 trillion, and equivalent to 12.8% of the GDP of 2020 and 2021 together).  The deficit was an astonishing 16.5% of GDP in the last 12 months of the Trump administration, and peaked at 19.2% of GDP in the 12 months leading up to March 2021.  It then came down rapidly, reaching a trough of 3.9% of GDP in the 12 months leading up to July 2022.  But it then turned upward, and is now at about 8.5% of GDP.

The figures are shown in the chart above.  They were calculated from the regular Monthly Treasury Statement released by the US Treasury, which has monthly figures on federal government receipts (revenues collected), outlays (expenditures in a broad sense), and the resulting deficit.  The most recent such statement (the one used here) was released in mid-August with figures through July.  The figures once published are rarely changed, and thus appear to be actual revenue and expenditure numbers and not estimates of what they were in any given month.

The figures will always bounce around substantially from month to month, due to factors such as when major payments on income taxes are due (e.g. each April), when expenditures are bunched (due to the fiscal year cycle), and other such seasonal factors.  Thus for the chart here I have calculated 12-month rolling totals for the figures, ending on the dates shown.  And I have expressed them as a share of average GDP over the period.  Since GDP figures are only available on a quarterly basis, I estimated month-to-month GDP figures based on an assumed constant rate of growth over the three months within each quarter.  The GDP figures were downloaded from FRED.

I also extrapolated the figures to the end of fiscal year 2023 – i.e. through to September 2023.  I suspect there will be a good deal of discussion on the sharp growth in the federal fiscal deficit in FY2023 when the full fiscal year figures are released in early October.  This post can be considered a preview, where while the final numbers will not be exactly the same as those estimated here, they will likely be close.

For the extrapolation, I assumed that federal outlays, receipts, and resulting deficits in August and September 2023 will be the same as they were in August and September 2022.  This was more reasonable than extrapolating the recent trend as the monthly figures fluctuate sharply due to seasonal factors, as noted above.  But it may well underestimate what the deficits will be in August and September 2023 due to the underlying upward trend of the past year.

For GDP for the third quarter of 2023 (where the preliminary estimate will not be released until near the end of October), I used the most recent forecast of third-quarter GDP growth produced by the Atlanta Fed.   The Atlanta Fed’s “GDPNow” forecasts have generally been quite good (far better than various consensus forecasts of panels of economists), and are based on a mechanical method where the forecast is first produced and then updated in real-time when key data are released – during the course of the quarter – on elements of what goes into GDP.  Correlations were worked out based on historical data, with those correlations then used – every time new data is released – to update the forecast of what GDP will be when an estimate is ultimately provided by the BEA at the Department of Commerce.

I have gone into a bit of detail on the Atlanta Fed’s GDPNow forecasting process as its most recent forecast (as I write this) is for GDP growth in the third quarter of 2023 to be quite high – at an annualized real growth rate of 5.9%.  This is a good deal higher than the most recent official (BEA) estimates of GDP growth of 2.4% in the second quarter of 2023 and 2.0% in the first quarter.  It is also substantially higher than the “Blue Chip” consensus forecast of a panel of economists of just 1.6%.  We will see who ends up being closer to the final figure on GDP growth, but for the chart above I used the 5.9% rate.  With this relatively high rate of growth for GDP, the federal deficit and other figures as a share of GDP will be biased in the downward direction.  Despite this bias (as well as that following from the use of 2022 figures for August and September, despite the upward trend in the deficit this year), the resulting federal fiscal deficit for FY2023 (which ends in September 2023) is conservatively estimated to be 8.5% of GDP.

A fiscal deficit of 8.5% of GDP in a period when unemployment is low is huge.  The unemployment rate has been at 3.7% or less (and as low as 3.4% – with this all within the range of statistical uncertainty) since March 2022.  It has not been at a level so low for such an extended period since 1968/69 – more than a half-century ago.  The increase in federal expenditures certainly in part accounts for this (Keynes is once again shown to have been right), but a fiscal deficit of 8.5% of GDP is not sustainable.

The gross federal debt held by the public (the relevant concept, as intragovernmental holdings of public debt will net out) was 95% of GDP as of 2022.  Round this up to 100% of GDP.  It is then easy to see that if one assumes, going forward, that real GDP growth will average 2.0% per year, say (it averaged 1.93% per year from 2000 to 2023), and if the inflation rate (the GDP deflator) matches the 2.0% Fed target for general consumer inflation, then the public debt to GDP ratio will remain flat if the fiscal deficit equals 4.04% per year (as that equals the compounded effect of 2.0% real growth and 2.0% inflation).  A fiscal deficit of 8.5% of GDP is far above this.  At such a deficit, the debt to GDP ratio will grow.  At a deficit of 4% of GDP or less, then with 2% growth and 2% inflation the debt to GDP ratio will fall from where it is now.

Why has the fiscal deficit grown by so much since the trough at 3.9% of GDP in July 2022?  I do not know enough about the fiscal accounts to say, but a few points can be made.  First, it was not due to rising interest rates.  While the US Treasury will need to pay higher interest rates on newly issued debt as the Fed has been raising interest rates, most Treasury debt is longer term and only comes up for renewal slowly over time.  According to the July 2023 Monthly Treasury Report, gross interest payments rose by $136 billion in the fiscal year to date (i.e. from October 2022 through to July 2023) compared to the same period in the prior fiscal year.  This is just 0.6% of GDP if one extrapolates it out to a full 12-month period,

In terms of the broad categories shown on the chart above, what was far more important was a fall in federal receipts (revenues) over the last year.  Federal receipts came to 19.6% of GDP in the 12 months leading up to July 2022, and fell to 16.8% of GDP in the (forecast) 12 months leading up to September 2023.  This was a reduction in receipts of 2.8% of GDP, and accounts for about 60% of the increase in the deficit during this period (which went from 3.9% of GDP to a forecast 8.5%, an increase of 4.6% of GDP).  An increase in federal outlays, rising from 23.5% of GDP in the 12 months leading up to July 2022 to 25.3% in the 12 months leading up to September 2023 (an increase of 1.8% of GDP), accounts for the other 40%.

I am not sure why federal receipts fell over the last year, but a guess would be that they rose in the period from early 2021 to mid-2022 (keeping in mind that these will always be for trailing 12-month periods) due to rebound effects from the Covid relief programs.  Those programs included deferral of when taxes would be due, and also provided for higher government expenditures on a variety of Covid-related activities.  These would translate into higher incomes, with this then leading to higher taxes later becoming due.  There was also major direct income support, although most of this was not taxable.

But for whatever reason, federal government receipts as a share of GDP have fallen substantially since mid-2022 and account for the major share (60%) of the increase in the deficit.  However, it is also important to note that while federal receipts have fallen relative to mid-2022, they are now (as a share of GDP) roughly where they were in early 2020, prior to the onset of the Covid crisis.  In fact, they are a bit higher, at 16.8% of GDP now compared to 16.5% in the 12 months leading up to early 2020.

Federal outlays are, however, substantially higher than where they were in early 2020.  They are now at 25.3% of GDP, versus 21.4% of GDP in early 2020, an increase of 3.9% of GDP.  This more than accounts for the increase in the fiscal deficit since then – rising from 5.0% of GDP in early 2020 (during the Trump administration but pre-Covid) to 8.5% now.  I have discussed before why the deficit of 5% of GDP during the Trump years (pre-Covid) was unwise, exceptional for a period when the economy was at full employment, and not sustainable.  The same is true with a deficit of 8.5% of GDP during a period when the economy is also at full employment.

Much of the new expenditures of the Biden administration are certainly high priority.  Climate change needs to be addressed, for example, and the US has long neglected its public infrastructure and there is an urgent need to repair it.  But I do not have a detailed breakdown of the expenditures, how they compare to expenditures before, and the needs being addressed.  But regardless, the now Republican-controlled House of Representatives will certainly force major cuts in federal expenditures, just as the Republican-controlled House elected in 2010 forced through major expenditure cuts during the Obama presidency.  Fortunately, the economy is now at full employment, while unemployment was still high in 2011 when the Republicans started to force major expenditure cuts on the Obama administration.  There had not been such cuts in fiscal expenditures with unemployment still high in periods recovering from a recession since before World War II.  The result was the exceptionally slow pace of employment growth following the financial and economic collapse of 2008/2009.

Attention should also be given to increasing fiscal revenues as part of a program to reduce the fiscal deficit.  A return of federal receipts to the approximately 18% of GDP share towards the end of the Obama presidency in 2016 would help in reducing the deficit to a sustainable level.  A good start would be to reverse the Trump tax cuts rammed through the Republican-controlled Congress in late 2017 on party-line votes – tax cuts that led to corporate income taxes being reduced by half and which introduced numerous new tax loopholes and other measures favoring the rich.  But with Republicans now in control of the House, that will of course not happen.

A deficit of 8.5% of GDP is not sustainable.  It will need to come down, with a deficit of 4% of GDP or less a reasonable goal.  There is a need for a constructive debate on how best to do this.  Unfortunately, given the state of American politics, it is unlikely that any debate will be constructive.

=========================================================

Update:  October 31, 2023

Final figures have now been released for FY2023 federal government outlays and revenues, with the release of the September 2023 Monthly Treasury Statement. The initial estimate of GDP in the third quarter of 2023 has also now been provided.  From these, we can calculate what the federal fiscal deficit was as a share of GDP in FY2023, and compare that to the estimates made when only data through July 2023 were available.

The FY2023 fiscal deficit came in at 6.3% of GDP.  This deficit figure is still high – it is well above the 5% of GDP rule of thumb on when one should be concerned (in an economy at full employment), and above the figure of around 4% of GDP where the government debt to GDP ratio would be flat.

But the 6.3% of GDP deficit is substantially better than the 8.5% estimate provided in the post above.  That figure was based on extrapolations from data that, at the time, were only available through July 2023.  Why the difference?  It turns out to be due almost entirely due to substantially lower than expected fiscal outlays in the final two months of the fiscal year (i.e. August and September).  The estimates made in the post assumed that the fiscal deficit in August and September 2023 would be the same as the deficit in those two months in 2022.  But the deficit turned out to be 2.1% of annual GDP lower in 2023 than it was in those same two months in 2022.  Within round-off, this is the same as the 2.2% of GDP difference between the 8.5% of GDP forecast, and the 6.3% of GDP realization.

This was basically entirely due to lower federal expenditures (outlays).  Those expenditures were 2.3% less, in terms of annual GDP, in August and September 2023 than they were in those two months in 2022.  Federal revenues were also a bit less, but only by 0.2% of annual GDP.

Note that all these figures have been presented in terms of annual GDP for the fiscal years 2022 or 2023.  But GDP is a flow, and GDP in just the two months being considered will of course be far less.  Based on the advance estimate of GDP in the third quarter of 2023, and assuming GDP in the two months will be two-thirds of that in the three months in the quarter (and also taking GDP at the quarterly, not annual, rate), the reduction in the deficit in those two months relative to the same two months in 2022 comes to 12.3% of the GDP of those two months.  That is gigantic!  While there is substantial volatility in the monthly figures, it is surprising that fiscal expenditures would change by so much in that period compared to what it was in the same period of the prior year.