Introducing
in:Our Authors
Joel Prakken is former Chief US Economist of S&P Global and IHS Markit, co-founder of Macroeconomic Advisers, and past president and director of the National Association for Business Economics. He has served as an outside advisor to the Congressional Budget Office, on the Advisory Panel of the Bureau of Economic Analysis, and as a consultant to the Joint Committee on Taxation. He holds a BA in economics from Princeton University and a PhD in economics from Washington University in Saint Louis.
President Trump has asked Congress to suspend federal fuels taxes temporarily in order to lower prices paid by drivers at the pump. Unfortunately, doing so probably won't produce the desired result, but energy companies certainly won't object. Here’s why.
A complete version of this commentary is available here.
by:Joel Prakken
|in:Viewpoints
The report from the Bureau of Economic Analysis on the “core” inflation rate for personal consumption expenditures (PCE) in March was concerning. The monthly change in that price index was 0.4% for the third consecutive month, for a 3-month annualized change of 4.4%! The 12-month change climbed from a recent low of 2.7% in October to 3.2% in March, well above the Fed’s 2% objective. And don’t forget: the 12-month change understates inflation, given how the Bureau of Labor Statistics (BLS) treated shelter costs last October when a partial government shutdown prevented the agency from conducting its monthly survey of consumer prices.
A refresher. Lacking data for many items, BLS assumed their prices in October were unchanged from September. For most items this understatement was corrected the following month when November’s prices were correctly recorded, except for the price index for shelter costs.
BLS calculates the monthly percent change in shelter costs as the 6th root of the percent change in shelter costs over the previous six months reported for one of six rotating panels within a larger sample of housing units. So, lacking survey data for October, the BLS assumed that shelter costs in October were the same as in April – the last time the panel scheduled to be surveyed in October was in fact surveyed – and then calculated October’s change in shelter costs as the 6th root of 0…equals 0! This understatement won’t be corrected until that panel is surveyed again in April. Until then, any change in shelter costs calculated over a span that includes October is missing a month of shelter cost inflation. If that span is less than a year but the change is annualized, then the understatement is annualized as well.
Let’s put numbers to this. The 6-month change in shelter costs was not recorded in October, but it was 1.65% (not annualized) in September. Let’s take that as an estimate of the true 6-month change in shelter costs for the panel that would have been surveyed had the government not been shut down. This implies that the price index for shelter costs has been low since October by 1.10651/6 – 1 ≈ 0.3%. Shelter costs have a relative importance of 35% in the consumer price index (CPI), implying that the CPI currently is low by 0.35*0.1% ≈ 0.1%, as is the 12-month percent change in the CPI.
On May 12 the BLS will release the CPI for April. No doubt it will show pronounced effects of higher energy prices, both direct and indirect, resulting from the closure of the Strait of Hormuz. However, it will also include the correction of the understated level of shelter costs. That correction will add approximately 0.1% to both the monthly and the 12-month percent change in the CPI, and slightly more than that to the corresponding measures of core CPI inflation. The impact on the price index for core PCE is roughly half this, given the smaller weight of housing in PCE than in the items covered by the CPI. So not only will the correction for the understatement of shelter costs boost reported inflation, it also will push the monthly change in the CPI above that for the PCE price index. None of this is earth shattering, but it is another reason to expect May’s inflation numbers to be unfavorable.
by:Joel Prakken
|in:Viewpoints
Global| Apr 08 2026On February 28 the US and Israel launched airstrikes on Iran. In retaliation, Iran closed the Straits of Hormuz. By early April the average US price of regular gasoline jumped to $4.11/gallon from $2.93/gallon in February, an increase of 40%.
To put this price shock in historical perspective, I: defined the real price of gasoline as the price index for personal consumption expenditures (PCE) on gasoline (and other motor fuels) over the price index for total PCE; calculated monthly percent changes in this real price of gasoline; weighted each of those changes by the geometric average of the current and lagged monthly shares of nominal gasoline purchases in total nominal PCE; cumulated the weighted price changes from January of 1960 to the present.
by:Joel Prakken
|in:Viewpoints
In early 2025 the Trump Administration announced a base tariff of 10% on imported goods, additional steep “reciprocal tariffs” on goods imported from countries running large trade surpluses with the United States, and additional levies on specific imported commodities. The average tariff rate on all imports surged from approximately 2.5% to 10%. The increase would have been larger if not for the many “slips twixt cup and (dutiable) lip.” Indeed, it might surprise many that today only 40% of imported goods are subject to tariffs (Chart 1).
by:Joel Prakken
|in:Viewpoints
Because the federal government was shut down in October the Bureau of Labor Statistics (BLS) did not conduct a survey of consumer prices that month but then reported two-month changes (for September-November) in prices with the following release of the Consumer Price Index (CPI). At the time I, in a commentary here (“A Dodgy CPI Rent Reading for November,” December 20, 2025), and others viewed the reported sharp two-month deceleration in the shelter component of the CPI with suspicion. Subsequent methodological clarifications from the BLS confirmed those concerns.
In October, for price levels not surveyed, the BLS assumed (unreported) changes of zero from September to October. In principle, with price levels then correctly measured in November, the two-month changes reported for September-November are correct. It is as if (implied) price increases in November include catch up effects, but with one important exception: rent.
The BLS stratifies its full panel of housing units into six subpanels that are surveyed in rotation. Each month the BLS assumes the monthly change in the shelter component of the CPI equals the sixth root of the six-month change in rent reported for the currently surveyed subpanel. Since a survey was not conducted during the shutdown, the BLS assumed that the rent for the subpanel that would have been surveyed in October was the same as in April when that subpanel was last surveyed. Because in October the six-month change in rent was thus assumed to be zero so, also, was the one-month change (i.e., the sixth root of zero).
by:Joel Prakken
|in:Viewpoints
On December 18 the Bureau of Labor Statistics (BLS) released the Consumer Price Index (CPI) for November. In October, when the federal government was partially shut down, BLS did not conduct its survey of prices, leaving most of them unrecorded for that month. Therefore, rather than reporting the usual 1-month percent changes in prices for November, BLS reported 2-month percent changes instead. For example, from September to November, the core CPI advanced at a 1% annualized rate – a surprisingly benign reading that, if accurate, significantly improves the current narrative on inflation and strengthens the case for easier monetary policy.
Unfortunately, the potential impact of the shutdown on both the quality and timing of the November survey raises legitimate concerns about the reliability of its results. One particularly dodgy-looking element of the report was a quite sharp deceleration in the CPI for shelter, the 2-month annualized percent change of which dropped from 3.9% in September to just 1.1% in November (gold line in chart). An erroneous reading here could be of considerable importance given that shelter costs comprise nearly 18% of “core” personal consumption expenditures.
The CPI-shelter reflects rents on tenant- and owner-occupied housing units. Imputed rents on owner-occupied units are inferred from observed rents on nearby comparable tenant-occupied units. Because shelter costs reflect average rents, they are highly inertial, lagging well behind current (i.e., marginal) market conditions for two reasons. First, rental contracts typically are for one year, requiring twelve months for all contracts to reflect a change in market conditions. Second, the BLS rotates through a panel of renters over a period of six months, adding another half year to the lag between marginal and average rents.
However, these lags allow one empirically to relate the CPI-shelter to current and past new rental contracts. I did so by estimating a model that explains the CPI-shelter with current and lagged values of Zillow indices of observed newly contracted rents. These indices are available monthly through November. I then used that model to forecast the shelter costs for the months of October and November. The resulting projection of the 2-month change in the CPI-shelter is shown in the blue line in the chart.
The model suggests that from September through November the CPI-shelter grew at an annualized rate of 2.9%, 1.8 percentage points faster than the number published by BLS. To me, the projection seems more believable than the reported figure. Replacing the latter with the former raises the 2-month annualized change in the core CPI by approximately 0.3 percentage points, to 1.3% - still a good reading, but not as good. And, of course, all this makes one wonder about the reliability of estimates of other prices in the report. So, before concluding prematurely that inflation is quiescent, it makes sense for one to await additional months of readings.
by:Joel Prakken
|in:Viewpoints
After forty-two days, the longest (partial) shutdown of the federal government on record, which began October 1, is finally behind us. Forecasters have scurried to gauge the shutdown’s impact on real GDP. The exercise is not simple. With savings as a buffer, delayed income is not necessarily delayed spending. With inventories as a buffer, delayed spending is not necessarily delayed production. Some delayed spending and production can be made up later, even within the current quarter. The task is further complicated because effects of the shutdown are not identified separately in the source data compiled by the Bureau of Economic Analysis (BEA). Nor does the BEA make special adjustments to GDP for the shutdown…with one important exception.
In our National Accounts services produced by government are valued at cost, the variable component of which is the compensation of government employees. The BEA treats furloughed federal workers as if their real compensation is zeroed out. This makes it straightforward to compute the real value of government services forgone during the shutdown, as well as the associated negative contribution to fourth-quarter growth of GDP. With approximately 1/3 of federal civilian workers furloughed, each extra week of the shutdown reduced the contribution of government services to fourth-quarter annualized growth of GDP by approximately 0.14 percentage point. Hence, the reduction in government services during the six-week shutdown will subtract approximately 0.8 percentage point from GDP growth in the fourth quarter.
by:Joel Prakken
|in:Viewpoints
It is generally understood that Gross Domestic Product (GDP) does not include the value of imports. It is, however, less well appreciated that in the System of National Accounts tariffs are treated as domestic value added. That is, customs duties are included in GDP.
From the first to the second quarter of this year customs duties surged $170.7 billion, from $97 billion at an annual rate to $267.7 billion -- an annualized growth rate of 5700%! -- as new tariffs imposed by the Trump Administration took effect. All else equal, if that increase in tariffs had immediately passed through to prices faced by final demanders of domestic product, the annualized rate of change in the GDP price index would have spiked nearly 3 percentage points in the second quarter. As it happened, GDP inflation in the second quarter was tame at 2.1%. Obviously, there were tariff slips twixt cup and lip.
by:Joel Prakken
|in:Viewpoints
From January to August the average tariff rate on all imported goods increased 9 percentage points, from 2.3% to 11.3%, while the price of those imports, recorded by the Bureau of Labor Statistics (BLS) excluding tariffs, was essentially unchanged. Hence, the price of imports, including tariffs, rose 1.113/1.023 - 1 = 8.5%, reflecting the full amount of the increase in tariffs (Chart 1, solid lines). This suggests the cost of the tariffs has been borne entirely by American businesses and consumers, with none of that cost passed backwards to foreign suppliers.
by:Joel Prakken
|in:Viewpoints
...was the reaction in financial markets after the latest employment report showed that, including revisions, growth of establishment employment slowed to a crawl over the last three months. Investors, who before that report were convinced the Fed will maintain its policy rate in the range of 4¼% - 4½% at the September meeting of the FOMC, are now equally convinced the Committee will cut the rate by ¼ percentage point, a shift in sentiment re-enforced by dovish remarks delivered by Chair Powell at the Fed’s annual shindig in Jackson Hole.
While not religious about it, I do find the Taylor rule a useful framework for contemplating the short-run dilemma facing the Fed as it attempts to satisfy the dual mandate of low inflation and high employment. For a given “neutral” real (i.e., inflation-adjusted) federal funds rate – today we call it real “r-star” – the original rule showed the Fed: (1) raising (lowering) the nominal policy rate by 50 basis points for each percentage point that real GDP exceeds (falls short of) potential GDP; (2) raising (lowering) the policy rate by 150 basis points for each percentage point that inflation exceeds (falls short of) 2%. The value of real r-star itself is not observed directly, only inferred from empirical models. However, the Fed’s Summary of Economic Projections (SEP) reveals policymakers’ views on the matter. For example, the June SEP shows, in the long-run, inflation of 2% and a nominal funds rate of 3%, implying a value of real r-star of 1% - unchanged from December’s SEP.
Another familiar rule in macro is Okun’s law relating changes in the unemployment gap to changes in the output gap. Using a typical textbook version of Okun’s law, the Taylor rule can be restated to show the Fed lowering (raising) the funds rate by 100 basis points for every percentage point that the unemployment rate exceeds (falls short of) the full-employment rate. The current unemployment rate is 4.2%, slightly below CBO’s estimate (= 4.3%) of the “non-cyclical” unemployment rate, while the 12-month change in the core PCE price index is 2.8%. If real r-star is 1%, then current values of the unemployment rate and inflation imply an appropriate policy rate of 4.3%, exactly equal to the funds rate in July.
From this perspective the Fed should be in no hurry to ease policy and, by cutting rates, risks perpetuating inflation above the 2% objective. The argument for a rate cut is that continued slow growth of employment will soon lead to rising unemployment that, through the Phillips Curve, will reduce inflation risks associated with a one-off increase in tariffs, allowing the Fed preemptively to shift its focus to the employment half of the dual mandate.
This is a demand-side narrative that may prove out, but there is a supply-side story to tell here as well. With each employment report the BLS publishes monthly data on the labor force that tautologically is the product of population and the labor force participation rate. In Chart 1, the orange line depicts the recent 12-month percent change of this series. Unfortunately, it is contaminated by “population controls” introduced by the BLS each January. Fortunately, the BLS publishes alternative monthly estimates of employment by age and gender that smooth through the population controls. By multiplying the reported participation rate into this alternative series for population, I calculated a smoother version of the labor force (blue line). Over the last year, but especially in 2025, its growth has slowed from 1½% to zero as the participation rate has fallen, especially among foreign-born workers – the consequence of fear engendered by the Trump Administration’s policies on immigration. And this surely understates the actual deceleration in the labor force because estimates of recent population growth do not yet reflect the sharp reduction in immigration – especially border crossings - over the last year.
Why does this matter? Firstly, addressing the current clamor for a rate cut: growth of employment that is lethargic because it is constrained by slow secular growth of the labor force does not necessarily imply a cyclical rise in unemployment requiring rate cuts under the Taylor rule. Indeed, along a steady-state path with no growth in the labor force, I would expect monthly changes of employment to be zero with a constant unemployment rate. Secondly, however, it is standard macro that, all else equal, slower growth of the labor force is associated with a lower equilibrium real interest rate – i.e., a lower real r-star. The logic is that a deceleration in the labor force results in a relative excess of capital the puts downward pressure on the rate of return.
To get an empirical feel for this I fired up my macro system - which, on its supply side, is a Solow growth model – and calibrated it to the current economy assuming the real r-star of 1% shown in the recent SEPs. Then I reduced the annual growth rate of the labor force to zero from a baseline value of 0.7% (the most recent 10-year average) and held it there through 2028 before allowing a gradual recovery. With the Taylor rule maintaining the economy close to full employment, the system kicks out the inferred change in the real r-star. The results of this experiment are shown in Chart 2. They show the deceleration in the labor force is associated with a decline in real r-star that averages about ¼ percentage point through 2028. This could justify the Fed cutting the nominal policy rate even with the economy near full employment and inflation running above target.
None of this is precise. There are unobservable considerations and ceteris that may not be paribus. While in the minority, there are Fed watchers worried that by cutting rates now, the FOMC may validate inflation above 2%. It is a legitimate concern but, thinking about it from a supply-side perspective on the economy, I find that risk a little less concerning.
by:Joel Prakken
|in:Viewpoints
In Samuel Becket’s renowned play, Vladimir and Estragon wait for Godot, who never arrives. It’s like waiting for the inflation impact of recent tariff increases…or is it?
On April 2, “liberation day,” the Trump Administration announced a baseline 10% tariff on most imported goods and additional “reciprocal” tariffs, quickly delayed until August, on select countries. This was on top pf other new tariffs on vehicles, parts, steel, aluminum, and other select commodities. I estimate that by June these initiatives increased the import-weighted average of tariff rates on goods - call it the “statutory” tariff rate - by 17.5 percentage points, to 20%. At May’s volume of imports, that implies an increase in the “run rate” of customs duties of nearly $600 billion at an annual rate, a staggering rise. If passed fully forward, it would lift the aggregate price level by roughly 1.7%. Yet so far, despite economists’ warnings, any impact on inflation has been difficult to discern. Why?
One reason is that imports temporarily surged early this year as importers rushed to beat the anticipated increase in tariffs (Chart 1). Until the resulting excess stocks are worked down, merchants can postpone selling goods subject to the new tariffs, delaying price pressures arising from the recent increases. But is there even more to it than that?
by:Joel Prakken
|in:Viewpoints
Before the 2024 presidential election my uncontroversial view of the near-term economic outlook was that real GP would grow near 2% for the next few years with the economy remaining near full employment, inflation subsiding towards 2%, and the Fed gradually cutting its policy rate. I assumed the personal provisions of the Tax Cuts and Jobs Act (TCJA) would be extended beyond 2025, and that both the limitations on state and local tax (SALT) deductions and the temporary business provisions in TCJA would sunset as scheduled under current law.
Now Congress is debating the One Big Beautiful Bill Act (OBBBA) while courts decide the legality of new tariffs imposed by the Trump Administration. Here I offer thoughts on how, directionally, these policies, if enacted, would shift my view of the near-term outlook for real GDP growth. To organize my discussion, I’ll group the policies like this:
• OBBBA
o Extend TCJA
o New tax cuts
o New tax increases
o Increased spending by the Department of Homeland Security
o Spending cuts, including Medicaid
• New tariffs
Let’s begin by asking how much failure to extend TCJA might undermine near-term growth. I’m skeptical of estimates suggesting the impact to be large but, given limitations here on space, a picture is worth thousands of words. The nearby chart shows annual real GDP growth from 2011 through 2024, including the first two years (2018 and 2019) when TCJA was in effect, but before COVID punctuated the near-term outlook. Then, like today, the economy was near full employment with inflation near the Fed’s 2% target. I don’t see a significant pickup in growth during those two years even though, as its centerpiece, TCJA cut the corporate tax rate to 21% permanently. Perhaps not all ceteris are paribus here, but would I expect failure to extend the other provisions of TCJA to have a big negative impact on near-term growth? No and in any event, as mentioned above, I expected the personal provisions of TCJA to be extended.
To help finance the reduction in the corporate tax rate, TCJA made other business tax cuts temporary and also included subsequent business tax increases. For example, under current law “bonus depreciation,” which has fallen from 100% in 2018 to 30% in 2025, ends next year. Limitations on deductions for interest and depreciation of R&D expenditures were implemented in 2022 and 2024. Here OBBBA would not extend current policy but rather would revive the initial TCJA provisions for five years. It also makes the TCJA treatment of certain foreign earnings permanent. I’d not assumed these “extensions” of the business provisions of TCJA, so I consider them new tax breaks that would boost my forecast.
by:Joel Prakken
|in:Viewpoints