What Attempts To Exclude Government Spending From Economic Data Tell Investors (May 20, 2024)

If the robust U.S. economy is a debt mirage, expectations for higher long-term interest rates are folly

By Lisa Beilfuss
May 20, 2024

Explosive government debt isn’t a new or undercovered story. But here is a statistic that recently caught our attention: Every dollar of fourth-quarter U.S. GDP growth was bought with $2.50 in new debt issued by the U.S. government.

The stat is from the 2024 In Gold We Trust report by Ronald-Peter Stoferle and Mark Valek of Incrementum AG. As they put it, the U.S. economy is being pushed to peak growth levels largely by the use of “economic steroids.” 

Boiling the U.S. fiscal situation down in this simple way makes the problem more palpable. This week we dug into two veins of related questions to help investors better evaluate economic data and anticipate the direction of policy. First, what does the U.S. economy look like when the government is excluded from growth metrics? Second, how realistic are widespread assumptions about higher long-term interest rates?

It isn’t possible to fully back out the impact of government spending on gross domestic product. But Peter Earle and Thomas Savidge of the American Institute for Economic Research have attempted to untangle the government from the private sector. Last month they published an analysis of economic performance “without government spending muddying the water.”

Using data from the Bureau of Economic Analysis, Earle and Savidge provide new calculations of gross domestic private product, or GDPP. This measure is obtained by subtracting government consumption expenditures and government gross investment from GDP, leaving personal consumption expenditures, business investment and net exports. The approach is imperfect because it doesn’t exclude transfer payments such as Social Security or unemployment insurance, as those payments are counted as private spending. Nor does it consider the impact of government debt, which weighs on economic growth, or the impact of historically expansive monetary policy.  

Even so, the analysis shows GDPP has slumped and the gap between GDP and GDPP has widened. AIEA finds the average annual growth rate of GDPP prior to 2000 was 3.63% while the post-2000 average was 2.34%. GDPP has been growing 35% slower per year since 2000, Earle and Savidge say, and their results likely overstate reality given they can’t fully exclude the impact of government spending. 

Here is another way to look at the economy excluding government spending. The BEA last reported a personal savings rate of 3.2%. That is below a 12-month average of 4.3%, below a post-Covid average of 8.3%, and about half of the average savings rate in the five years before massive pandemic stimulus. Still, private survey data make the official savings rate from the BEA look high. Here are two examples: First, a 2023 survey by Payroll.org said 78% of Americans live paycheck to paycheck, up 6% from 2022. Second, a LendingClub report in late 2023 showed 62% of Americans live paycheck to paycheck, including more than half of households earning over $100,000 a year. 

Richard Farr, chief investment officer at Merion Capital Group, offers a way to square a positive, if falling, savings rate with conflicting private reports. The savings rate is based on disposable personal income. When the BEA calculates personal income, it adds federal entitlement spending. In other words, the savings rate reflects government transfers–not just earned income. 

“If income isn't actually income, then ‘savings’ can't be savings,” he says. If all entitlements such as Social Security, Medicare and Medicaid were backed out, Farr says the savings rate would fall from roughly positive 3% to roughly negative 7%. “Is our robust economy simply a debt mirage?” Farr asks. 

Debt mirage or not, it is unrealistic to expect politicians to dramatically cut spending anytime soon. Therein lies what is probably the answer to a separate question around prevailing interest-rate expectations.

Many economists argue that the long-run natural rate of interest—or the inflation-adjusted short-term interest rate when an economy is at full strength and inflation is stable–has risen significantly after being stuck around 0.5% for the past 15 years. That is the rate often referred to as r-star, and it is key to setting monetary policy and informing investment decisions. (If inflation runs at 2% a year, then you get what has been the Fed’s projection of 2.5% for the long-term Fed funds rate).

RSM chief economist Joe Brusuelas argues that r-star is now closer to 2% than 0.5%. If the inflation target remains at 2% and is respected, that translates to a 4% terminal Fed policy rate. Behind his estimate is a view that investments in infrastructure, domestic manufacturing capacity and environmental sustainability are creating the conditions to improve productivity and lift potential growth. The bottom line, says Brusuelas, is that the era of extremely low interest rates has ended. 

But the logic is circular, at least in part. Infrastructure investment and environmental sustainability have been particular targets of government spending. Now, there is a bipartisan push in the Senate to spend at least $32 billion over the next three years to develop AI, a key source of productivity growth expectations.

As David Rosenberg of Rosenberg Research notes, the Federal Reserve has the power to ease or exacerbate the federal government’s fiscal woes via its interpretation of r-star. He highlights the interest-expense-to-revenue ratio, what he calls the public sector version of the interest coverage ratio, as the key fiscal metric to evaluate what happens when you combine a higher for longer interest rate stance with high and rising public debt. 

“Put together high debt levels and high rates, and you get a very rapid deterioration in interest coverage. What’s more, it lasts; the government needs to supply assets at maturities along the full yield curve, so those higher expenses get locked in for years to come,” says Rosenberg. He notes that the period of high interest rates between roughly 1975 and 1985 caused the interest-expense-to-revenue ratio to double a decade later, even as the government was running surpluses and reducing the debt. 

Even partially backing out fiscal spending reveals slower growth and questionable productivity, suggesting the government will increasingly need to spend more to maintain output growth. Doing so crowds out private-sector growth and burdens younger generations, underpinning the vicious cycle. The era of low rates isn’t so much an era, then, but an ongoing political and economic reality. The irony is that for the economy to withstand higher long-term interest rates, the government must seriously address its profligacy.