Forecasting the economy right now feels a bit like trying to carve a path through thick jungle undergrowth on a foggy day. There are multiple layers of confusion and a forecast has to address these issues first before tracing out a possible path forward.
To be specific, at the start of 2026, there are three broad areas of confusion. First, there’s a huge divergence between dismal consumer confidence and a euphoric stock market, neither of which appear to reflect actual economic trends. Second, recent economic data are distorted and, in some cases, missing altogether, making it difficult to discern the starting point for any forecast. And third, even as economists struggle to assess the reverberating effects of the administration’s 2025 policy moves, new policy actions this year, along with Supreme Court decisions, could have a major bearing on the outlook.
Still, a forecast must start somewhere, so here is a brief summary of how we address these issues, what this implies for the near-term economic outlook and what this means for investors.
Dealing with Distortions
Consumer sentiment still seemed to be understating economic strength at the end of last year. A November unemployment rate of 4.6%, combined with year-over-year CPI inflation of 2.7%, yields a “misery index” of 7.3%, up from 6.9% a year earlier but still better, (that is to say, lower), than it has been 76% of the time over the past 50 years. Meanwhile, the University of Michigan Consumer Sentiment Index came in at 52.9 in December – worse than in 99% of the months over the past 48 years. Even adjusting for a methodological change to the survey in 2024, sentiment is lower than it has been 96% of the time since 1978.
I wrote about this last November, in a piece entitled “Why a 1:13 expansion feels like a 1:20 recession” and concluded that the gap between sentiment and broad measures of economic performance was likely due to three factors: the depressing impact of social media, greater political polarization and rising income and wealth inequality. This last effect is particularly relevant for surveys that weigh all respondents equally as opposed to other economic statistics, such as income and spending, that are far more impacted by the fortunes and behavior of the rich.
In sharp contrast to sentiment, the S&P500 provided a total return of 18% in 2025 following huge gains of 26% and 25% in the prior two years, respectively. But just as confidence readings are understating economic performance, stock market gains appear to exceed any realistic assessment of economic success. Part of this may relate to the circular nature of many transactions in the technology sector generating a profit surge that is, like the intelligence it is building, somewhat artificial. In addition, as I noted in an article last month entitled “Why Stocks are Outperforming the Economy”, the collective impact of structural changes such as increasing inequality, the rise of defined contribution retirement plans and embedded capital gains are pushing money into the stock market while deterring investors from cashing out. As noted by some readers of that article, I could also have included the decline in the number of publicly-listed companies and the increased use of corporate buybacks since the 1980s among these stock-price-enhancing structural trends.
Cleaning up the Economic Data
However, the bottom line is that actual economic data likely paint a truer picture of the economy than either Main Street or Wall Street perspectives. That being said, the economic data themselves have problems that need to be addressed.
The first immediate issue is that we are missing data for the October 2025 CPI and Household Survey Jobs reports. We address this, in broad terms, by interpolating the data between September and November and then running forecasting models through November, using fitted values to refine the October estimates. This isn’t exact but it shouldn’t seriously bias forecasts going forward.
There is a more serious problem, however, in the CPI survey relating to rents and owners’ equivalent rent which, combined, account for a third of the CPI basket. The BLS uses a rolling 6-month panel to assess rents and, without October data, they assumed no change in either rents or owners equivalent rent between April 2025 and October 2025, impacting one sixth of the panel. This issue will resolve itself with the release of April 2026 data in a few months, which will include a 12-month increase rather than a 6-month increase for that part of the panel. However, until then, year-over-year published inflation will underreport actual inflation.
On the employment side there are two major problems. First, the household survey normally has a once-a-year adjustment to population numbers in January, along with new population growth projections for the rest of the year. This included a massive bump-up in population in January 2025 reflecting a past surge in immigration. It also included forecasts of a solid increase in population in each month of 2025.
These forecasts are likely entirely wrong due to the implementation of a dramatic change in immigration policy. However, the Census Bureau has not provided the BLS with updated estimates on population for 2025, so there will be no population adjustment in the January jobs report, due out on February 6th. Until this issue is resolved, while the unemployment rate should still be valid, the household survey will be relatively worthless in estimating the raw numbers of people employed, unemployed or not in the labor market. In our forecast, we focus on getting the unemployment rate right and we just wait on better data to project raw numbers.
In the payroll survey, the preliminary benchmark revision estimate for March 2025 was -911,000. This estimate has not been folded into monthly numbers yet and will be revised before this occurs with the release of the January payroll employment numbers on February 6th. Not only is this very likely to cut estimated job growth in the year that ended in March 2025, but it will likely reduce job growth thereafter, an issue which we address when projecting monthly job gains for 2026 and 2027.
A Spoonful of Sugar and the Tariffs Go Down
Finally, there is the issue of federal government policy going forward. We assume that the federal government’s hard line on immigration continues, cutting net immigration to 250,000 per year, compared to roughly 1,000,000 per year in the 20 years before the pandemic.
On trade, we assume that the Supreme Court will overturn the administration’s so-called “reciprocal tariffs” by the middle of the year, and that the administration only partly restores the lost revenue by imposing tariffs using other legislation. We assume that this will reduce the effective tariff rate on goods from 11.0% entering 2026 to 7.5% by the end of the year, lowering inflation and boosting economic growth but also contributing to a deteriorating fiscal situation.
We also assume that the administration succeeds in pushing through “tariff rebate checks” by the middle of the year. Republicans still control Congress and will want to sustain consumer spending ahead of the November mid-term elections as the impact of a bumper crop of income tax refunds wanes. We assume that this will take the form of a $2,000 payment deposited into taxpayer checking accounts in the third quarter and that they are distributed to the roughly 75 million households currently earning under $100,000, thus adding $150 billion in fiscal stimulus.
We also think it likely that the Democrats will retake control of the House of Representatives and that this will result in much less policy change for the remainder of the President’s term.
A Baseline Forecast and Investment Implications
So, given these assumptions, what does a baseline economic forecast look like?
First, following less than 1% growth in the fourth quarter of 2025, we expect real GDP growth to average over 3% in the first three quarters of 2026, but then slow in the fourth quarter and into 2027. While growth could average 2.2% for 2026, without further fiscal stimulus, it could slip back to 1.7% in 2027.
On jobs, a lack of labor supply and strong productivity gains could limit payroll job gains to an average of 60,000 per month over the next two years. However, because of only limited growth in the labor force, the unemployment rate could still hold steady at close to 4.5% throughout 2026 and 2027.
On corporate profits, we expect continued high-single-digit gains in pro-forma S&P500 earnings as a continued capital spending surge by tech companies boosts revenues faster than depreciation and companies more broadly continue to do a good job in controlling costs and leveraging productivity gains.
On inflation, we expect CPI inflation to accelerate to 3.6% year-over-year by June of this year, boosted by tariff feedthrough effects and fiscal stimulus. However, we expect that lower tariff rates and fading fiscal stimulus in late 2026 cause CPI inflation to fall to 2.2% year-over year by the fourth quarter and to hover close to that level into 2027, allowing the Fed to finally achieve its 2% consumption deflator objective
With no real recession threat, elevated inflation in the near term and the prospect of further fiscal stimulus, we expect the Federal Reserve to take its time in implementing further rate cuts with two in total in 2026 and a further cut in 2027. We believe this will be the case even with new Fed leadership, as a majority of Federal Reserve governors and regional bank presidents appear very committed to maintaining Fed independence.
That being said, we still expect the Fed to ease more than the ECB with the Bank of Japan raising rates in a relatively stable global growth environment. This should allow the dollar to resume its decline, albeit at a slower pace than in early 2025.
For investors, this may all seem boring and familiar: moderate growth, moderating inflation, stable unemployment and solid profit gains. Barring shocks, this could sustain the 2025 equity market rally into 2026 and beyond. However, the expansion could be derailed by political or geopolitical shocks. Long-term interest rates could rise if investors fear a loss of Fed independence or even faster-rising federal debt. In addition, U.S. equity valuations look high, making large cap stocks, particularly those that have been bid up in AI euphoria, vulnerable. Finally, after three years of huge stock market gains, many investors, while richer than they were a few years ago, also have more concentrated and risky portfolios than they probably intended.
This being the case, while just figuring out a baseline forecast is not easy, investors should try to look beyond it and should devote at least as much attention to what could happen to their money if 2026 turns out to be much more dramatic, in either direction, than assumed in a middle path for the economy.