Modeling Tariff Impacts: The Directions are Clear but the Magnitudes and Persistence are Not
- High quality analysis from independent think tanks suggests that for the current constellation of tariffs (25% on Mexico and Canada, and 20% on China) growth will be roughly 50bps and 10-20bps lower in 2025and 2026 respectively and paired with cumulative inflation of 1-1.25% in the next few years. At least that’s a reasonable baseline if they remain in place.
- The level impacts on GDP tend to partially fade over time as trade patterns adjust.
- The issue with these modeling exercises is the inherent difficulty in capturing the non-linear nature of sectoral downturns and recessions, which general equilibrium models like this are not designed for. In addition, second round inflation risks loom. The initial conditions for a stepped up trade war are worse than in the first Trump administration. Beyond the actual tariffs there is likely to be near-term uncertainty drag.
- One should keep extremely wide confidence intervals in place when forecasting the administration’s actions and the downstream macroeconomic impacts. Given the breadth of stated reasons and uses for tariffs, there are a large number of potential offramps available, but their timing and likelihood remains highly uncertain.
Given the implementation of 25% tariffs on Mexico and Canada and an additional 10% on China (taking the rate there to 20%), the possible risks are becoming reality. Even if these tariffs see some resolution and are ultimately removed, the administration’s current ‘everything, everywhere, all at once’ approach will not allow the general uncertainties to fade quickly. Now that we have moved from tariffs risks into realities, this is challenging many market participants’, and like C-suites’, underlying assumptions about the administration. It is certainly possible that the current tariffs will merely be a phase in a negotiation process, a downside blip 5y into the post-covid cycle, but that remains to be seen. For now, at least, they are policy.
Taking for the moment these tariffs as persistent fait accompli, most general equilibrium models find relatively modest impacts, all things considered. Modeling work from the Yale Budget Lab and the Peterson Institute for International Economics (admittedly with only 10% China tariffs) are helpful expert benchmarks. On growth, the YBL work finds a peak cumulative hit to the level of GDP of around 0.8-1%, depending on retaliation, that gradually fades to a ~35bp hit. The PIIE study finds somewhat smaller impacts, with a peak level drag of 30-60bps in 2027 that fades to roughly half that in the long run, but this is an understatement of current shocks given they only include 10% China tariffs. On inflation the results find that, with overseas retaliation, price levels rise 1-1.5% cumulatively over the next few years.
The clear risk is that these sorts of general equilibrium models fail to take into account shorter-run business cycle dynamics and some already beaten-up sectors, like autos and housing, may see larger impacts. The current somewhat confusing macro backdrop with few obvious private sector vulnerabilities and overextensions but also plenty of later cycle indicators makes it difficult to assess whether or not the economy would smooth through these dislocations (as in an environment of rapid robust growth) or accentuate them. Slowing immigration and labor force growth, as well as possible fading of some post-reopening productivity growth gains, are naturally lowering the trend pace of growth and the slowing of catch-up hiring and DOGE job cuts also raise labor market risks some. This will give the economy less cushion to absorb downside surprises while keeping growth positive.
In addition, a larger-than-baseline hit to growth could also lead to more persistent negative hysteresis effects which dent the level of GDP in the long-run above and beyond the general equilibrium results noted above. This is quite a speculative thought but both the GFC and post-covid experiences suggest that short-run demand shocks (really balance sheet and fiscal policy in the historical examples, perhaps an aggressive sudden unwinding of 30y of trade and supply chain assumptions could be an analogous, if smaller, shock to activity) can have surprisingly persistent impacts on the medium and longer-term.
On inflation, the clear risk is that while the textbooks would suggest that a supply shocks like these tariffs should be looked through by the Fed inflation expectations may not allow that to be feasible. Normally the Fed could respond largely to growth hit or at least take a patient inertial approach to see which impact wins out but the recent legacy of inflation, margin preservation despite supply chain shocks, and wage pressures raises the risks that more persistent second round impacts bind on the rate of inflation, not just the price level (in effect this is inflation hysteresis).
But as we have learned since covid the economy can often exhibit quite notable asynchronous behavior. If tariffs do persist, we may simply see consumer spending rotate back into services or something analogous to the 2014-15 mid-cycle industrials drawdown which has minimal broader impacts on the cycle.
The risks though are larger than they otherwise would be given a labor market whose endogenous 1-2y out direction of travel is not particularly clear and the legacies of high recent inflation.
