Biden’s Big Boom and Big Risk: 3 Scenarios

As I wrote in my last 2 posts, “Central Banks vs Bond Markets” and “Biden’s Stimulus Plan, Inflation, and Interest Rates”, there is a realistic risk of inflation and an interest rate shock for 2023, with potentially serious consequences for all markets and the real economy, and perhaps also politics. Lets explore 3 possible scenarios, with numbers.

Although inflation “predictions” are notoriously wrong and generally simply based on a slow-moving “auto-regressive process” (i.e. that inflation is most explained by past inflation), still, assumptions are what make the difference between scenarios. Other than past inflation, here are the variables that matter in building scenarios about how much inflation and how severe of an interest rate shock could be coming:

  1. The “fiscal multiplier” over a 24 month period (more on this later).
  2. Personal savings and spending behaviour by households.
  3. Estimated potential GDP, which goes hand-in-hand with the estimation of the output gap.
  4. Velocity of money.
  5. Monetary policy.

I will take for granted that vaccination and the pandemic will gradually improve all through 2021 and that the US economy will be roughly “back to normal” or almost by 2022. This seems like a reasonable hypothesis, but EVEN if things did not improve as much, it would not significantly change any of the 3 scenarios.

I will also take for granted no significant change in trade policies over the next 2 years, hence the protectionnist measures will stay in place.

I will suppose no major extra depreciation or appreciation of the USD in global markets going forward, but EVEN if there was, it also would not significantly change the general picture for the 3 scenarios.

Note that the current administration will tend to be more prone to environmental regulation, carbon taxing, and will tend to restrain low-cost fossil fuel energy supply, all of which are more prone to cost pressures and inflation. I am not saying this is “wrong”, but it will tend to add cost pressures to the mix. I will not take this into account, but it should be kept in mind.

(Full disclosure: I am a pro-Biden, pro-DEM, center-left progressive economist).

For the non-economists: the “fiscal multiplier” is the total short run (2 years) net impact on actual expenditures and production following a rise in government spending or a cut in taxation. Note that details matter A LOT with the fiscal multiplier (I can’t stress this enough!), but the general idea goes like this: the government spends 1 trillion (for example), which necessarily becomes 1 trillion income somewhere, for someone, hence those “someones” spend part of that 1 trillion, say 50% of it, so they spend 500 billion… and that 500 billion of spending is income for someone, who then spends part of that, and so on. In the end, the total “net hit” is more than the initial stimulus.

Fiscal multiplier: NBER and Fed research suggests estimates for the fiscal multiplier between 1 and 1.4 when the economy is below potential and monetary policy is unchanged. Note that the fiscal multiplier is higher when the economy is running below potential and lower when the economy is above potential (overheating). I will thus use 1.2 for my “baseline” (most realistic) scenario, 1 for my conservative (low inflation) scenario, 1.4 for my bold scenario. Note that the “bold scenario” will simply use normal assumptions, but I want the whole thing to be quite conservative, so that I am not accused of hyperbole. The fiscal multiplier now is likely high, as the stimulus is NOT in the form of tax cuts to upper middle class and high incomes who save all the extra money into assets, it is now sent to middle and low-income households.

Personal savings: personal savings exploded in the past year due to no possibility to spend and probably also precautionnary savings. The normal level for aggregate personal savings is about 1.2 trillion. A high level is 2 trillion. A VERY HIGH level is 3 trillion. It now stands at 4 trillion. For my 3 scenarios, I will thus use 3 trillion as my baseline (which means only 1 trillion out of the “extra 3 trillion” of the past year will be spent, a realistic but very conservative hypothesis), 4 trillion as my conservative scenario (no spending of current sky-high savings, very conservative), and 2 trillion total savings (still a high level), hence 2 trillion of spending as my “bold” scenario. People will start spending again in 2021–2022, as things normalize. Especially in 2022, with the virus essentially out of the picture and life going back to normal. Savings are unlikely to remain at 4 trillion.

Personal savings:

Estimated potential GDP: the low estimated potential GDP of the CBO is 20 trillion for late 2022. Note that the lower the potential GDP, the more inflationnary stimulus is, as the economy hits capacity constraints faster. It is generally accepted among competent economists that the CBO estimate is too conservative, hence I will not even use it in ANY scenario. I will use 22 trillion for ALL my scenarios, a full , hence WAY LESS inflationnary to the point of over optimistic (projecting low inflation). Real GDP is currently about 19 trillion, hence 3 trillion below my chosen potential.

Velocity of money: the speed at which money changes hands is pro-cyclical, meaning it contracts in recessions and increases in expansions. It crashed dramatically in 2020 and is likely to increase a bit in 2021–2022, which would be inflationnary at least at the margin, but to remain on the conservative side (projecting less inflation), I will suppose no change in the velocity of money for all scenarios.

Monetary policy: as per the Fed, I have written extensively about this in the last 2 posts, which I strongly encourage you to read here and here. But lets wrap it up this way: the Fed is promising zero change in policy for all of 2021–2022 and even 2023, and are saying it will remain strongly expansionnary. So they are likely to “print” more money to buy the “Biden Boom Bonds”, as well as start maturity rotations. But all this doesn’t matter.

Absent massive supply shocks, all major inflation periods generally happen when you have the COMBO of 1) debt-financed government spending combined to 2) significant M2 money stock expansion (not just the monetary BASE, but really M2) way above historical norm, and 3) a positive output gap (overheating economy). This is the “inflationnary trio” that is seen in all inflationnary periods.

Year-on-year M2 growth, 1960–2021:

Reminder: monetary and fiscal expansions are inflationnary , as I explained in my last post with this simple graph:

Stimulus facts:

Trump stimulus of just 3 months ago: 900 billion

Biden stimulus just signed now: 1900 billion

Total government stimulus: 2.8 trillion stimulus total that will have an impact in 2021 and 2022. We are now ready for the 3 scenarios:

Scenario 1: lowest inflation projection

Fiscal multiplier = 1, hence net hit from stimulus = 1*2.8 = 2.8 trillion net stimulus impact.

Spending of households of 0 trillion (personal savings remains at the very high level of 4 trillion and none is spent).

Potential GDP = 22 and GDP = 19 now, so output gap = -3 trillion.

Projected output gap for late 2022 : 19 +2.8+0–22 = -0.2 trillion (-1%), hence no inflationnary pressure, with normal inflation and bond yields of around 2% or 3%.

Scenario 2: baseline (realistic inflation projection)

Fiscal multiplier = 1.2, hence net hit from stimulus is 1.2*2.8 = 3.36 trillion net stimulus impact.

Spending of households of 1 trillion (personal savings will go from 4 trillion to 3 trillion, which is still very high savings relative to the historical norm).

Potential GDP = 22 and GDP = 19 now, so output gap = -3 trillion.

Projected output gap for late 2022 : 19+3.36+1-22 = +1,36 trillion (+6%, which is a very high output gap, but whatever, those are the numbers), hence perhaps 7% inflation and a bond yield around 5% by late 2023.

Scenario 3: bold-yet-still-possible scenario (highest inflation projection)

Fiscal multiplier = 1.4, hence net hit from stimulus is 1.4*2.8 = 3.92 trillion net stimulus impact.

Spending of households of 2 trillion (personal savings will go from 4 trillion to 2 trillion, still higher than the historical norm).

Potential GDP = 22 and GDP = 19 now, so output gap = -3 trillion.

Projected output gap for late 2022 : 19+3.92+2–22 = +2,92 trillion (+13%, which is so high that it is not really possible even remotely), hence at least 10% inflation and a bond yield around 10% as well.

Several scenarios

Supposing the same high potential GDP of 22 trillion for end of 2022, here are a few scenarios in the same reasoning as above:

Everything hinges on these 4 questions / assumptions:

  1. What is potential GDP, hence what is the output gap now?
  2. What is the fiscal multiplier?
  3. What will households spend of their huge stockpiles of savings?
  4. What is the inflationnary impact of a given positive output gap for a 30% M2 growth, which is more than double the peak of the last 70 years?

Although all of these are important to pin down and clarify, question 4 is THE question that needs to be explored and get some clarification. Indeed, inflation seems to , other than major supply shocks, so maybe we could have a very positive output gap with muted inflation. This is a possibility. Maybe the unemployment rate could go to 2% and we would only have about 3% or 4% inflation, which is still quite tolerable. If such is the case, then my “baseline” would call for 4% inflation and a 10y bond yield of about 3% for late 2023.

Typically, inflation rises about one year after peak output gap. This happened after the peak positive gaps of 1973Q2, 1978Q4, 1989Q3, 2000Q2, 2006Q1, and 2019Q4: inflation peaked about one year after hitting peak output gaps. All these happened with significantly less M2 growth, even in the 1970s (although the 1970s inflation was largely due to oil price shocks). That makes my baseline scenario at least plausible and realistic. So if we hit peak output gap of +6.2% in 2023Q2 of the “baseline”/conservative scenario, we would get lots of inflation, an interest rate shock, crashing markets in 2024, and a major recession in the real economy in 2024 or 2025, during or right after the election year… The crash in markets COULD be earlier, if all that I am saying here starts to be priced in and forward-loaded in late 2022, if inflation starts rising clearly. We shall see.

Output gap (blue) and inflation (red), 1950–2020:


My baseline is for about 7% inflation somewhere in 2023 or 2024, 10y bond yields around 5%. This has severe consequences. Asset prices and real estate prices are sky-high and will remain on the uptick for some time. Inequality and social/political divisions are sky-high. An interest rate shock of this magnitude would bring a major crash in markets and the real economy. Markets crashed with a very mild inflation and monetary tightening in 2018. An inflation and bond yield shock (hence interest rate shock) of several percentage points in 2023–2024 would severely destabilize markets and the real economy.

Given the consequences of a monetary tightening in the current social, economic, financial, and political context, the Fed’s promise to keep inflation will certainly be challenged. If the shit hits the fan and they have to choose between a major jump in bond yields and tightening policy OR printing money and buying bonds to keep bond yields low (and the hell with inflation), what will they do? Implicit fiscal dominance.

The mystery

These are all cold-facts calculations. The Fed and most serious economists are projecting anywhere between 2% and 4% inflation. My very cold-facts baseline estimate is for 7% inflation and 5% bond yields by late 2023.

What am I not getting? Somebody tell me. I know that “central bank credibility” can keep inflation expectations “anchored” and hence also inflation and interest rates lower than they otherwise would be, but can it really make inflation be 3% when a realistic projection is for 7% inflation?

Perhaps the Fed and others are supposing things that I consider unlikely: that households will NOT spend their sky-high savings of 4 trillion while the “normal level” is 1 trillion and a “high level” is 2 trillion; that potential GDP is indeed as much as 10% higher than the official measures (this is my assumption all along in this text); that the fiscal multiplier is close to 1 and not 1.2 as I suppose here; that the aggregate supply curve is very flat (alternatively that the Phillips Curve is flat), even when GDP goes above potential and unemployment goes to 2%. This would mean we could print a LOT without inflation.

Pascal Bedard

This was on a rock climbing trip in Red River Gorge, Kentucky, USA

Sharing thoughts on economics, finance, business, trading, and life lessons. Founder of

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store