Market Insights

Should You Believe the Atlanta Fed?

March 21, 2025
Bipan Rai headshot

Bipan Rai

Managing Director, Head of ETF & Structured Solutions Strategy

Market chart showing recession risk

Topics covered:

  • Why US growth in Q1 could be negative
  • CAD CPI should remain sticky in the coming months
  • Portfolio strategy updates

1.) Portfolio Strategy:

Earlier this week, we flagged ZDI and ZMT as funds to watch in the period ahead. 

We should also add ZGI to the mix. That’s for the simple reason that we anticipate ex-North America fiscal policy to be loosened and for countries to start focusing less on external trade and more on domestic investment.

We’ve seen Germany make inroads there, and we suspect other countries will follow.

2.) US Notes:

a.) The Atlanta Fed GDPNow tracker is signalling that the US economy likely contracted by 1.8% (annualized) in Q1 this year.

  • The largest contributions are expected to come from net exports (imports likely exploded higher as businesses tried to source inputs sooner) as well goods consumption (as tariffs increased prices).
  • Net exports are still the key factor to watch. Stripping those out – alongside inventory investment – indicates that sales likely expanded by +1.2%.
  • That’s not a bad figure for one quarter, but it’s an early sign that momentum in the US economy is slowing once we look past net trade and inventory investment. Recall that same series has averaged +2.7% y/​y over the past three years.
  • A loss of momentum is also in sympatico with our view that the current phase of expansion in the US feels mature.

But can you trust the Atlanta Fed GDPNow model? Over the past 10 years, the average tracking error just prior to the release has been around -0.32%. On absolute terms (which means we don’t care about the direction of the miss), the average over that same time frame is 0.79%. Keep in mind this includes the insane tracking errors between Q1 2020 and Q2 2021 – which corresponded to peak forecasting uncertainty thanks to Covid-19.

We’ve still got another month-and-a-half before the advance reading for Q1 is published. But even still, based on its prior history, I wouldn’t dismiss the scale of the negative print that we’re getting from the Atlanta Fed GDPNow model at all.

b.) Trump firing both Democrat commissioners at the FTC opens up a very controversial issue that could impact markets. That being, does he have the ability to make changes at other independent agencies – namely the Federal Reserve.

Legal precedent is fairly clear that he doesn’t, but we’ve also seen more rhetoric from the White House aimed at challenging the judicial body.

Institutional stability is an integral part of why US markets draw capital from elsewhere. A fraying” of that stability leaves US markets vulnerable on a go-forward basis.

3.) FOMC Meeting:

a.) The base case for today calls for the Fed to remain on hold and make token changes to the statement. Meanwhile, QT cessation will be discussed and likely actioned into April – with a slight chance that both steps are condensed into today.

There’s not much priced into the USD OIS curve for today. Instead, markets are of the view that any potential rate cuts will be a story for the June-July-Sept FOMC dates.

b.) Shortly before the Fed entered its blackout period, we had heard from several FOMC members that were in the neutral/​dovish camp (Powell, Kugler, Daly) and the messaging was fairly straightforward:

  • Given trade, immigration, and fiscal policy right now – there’s far too much uncertainty.
  • That uncertainty is a near-term source of demand constraint for domestic activity.
  • Despite the above, longer-term inflation expectations are still stable and consistent with the Fed reaching its 2% target in due time.

That last point suggests that the costs to being cautious are low. As such, keeping rates on hold while monitoring the trade backdrop and how that filters through to the real economy makes sense at this point.

In terms of messaging and guidance, we probably won’t get much from the Fed statement or from Chair Powell’s presser. That is largely because the Fed needs that flexibility as the macro backdrop remains incredibly challenging to navigate. Therefore, we doubt there’s any meaningful shift in key verbiage from prior statements (i.e – risks to employment and inflation are still roughly in balance), while an acknowledgement is made towards the economic outlook being highly” uncertain.

c.) However, this is a forecasting round for the Fed, and the greatest market moving potential comes from where FOMC participants project the median dots for the coming years to be. 

In December, the 2025 dot was at 3.875%, which implied that the FOMC saw two cuts for this year. In order to shift that dot lower (implying more than 2 cuts for this year), we need to see at least five members that voted for 3.875% (or higher) in December to shift their projection below that level. Alternatively, in order to shift from 2 cuts to fewer this year, six FOMC voters would need to shift their projection from 3.875% (or lower) in December to above that level now. Needless to say, the bar feels high for a change in either direction.

d.) For what its worth, futures markets (SOFR) are currently pricing in just over two cuts by the end of this year. The general feel is that the risks to that are skewed towards more cuts given the drops in business and household confidence of late (see the Michigan number from last Friday) as well as what the uncertain trade backdrop means for US growth.

However, as we’ve mentioned above, the Fed’s messaging will be deliberately vague this week. That means that we can probably still look for a better entry point for steepeners or ZTS from here.

4.) Canada News + Notes:

a.) The magnitude of the CPI print was a bit surprising to us – with the headline coming in at a whopping +1.1% m/​m gain. That corresponds to +2.6% y/​y (from +1.9% last month).

Of course, the easy explanation here is that GST/HST holiday ended in the middle of the month (as taxes paid are included as part of the CPI calculation). But even if we strip out the effects of direct taxes, both of the BoC’s preferred gauges (CPI-Trim and CPI-Median) expanded by +0.3% on the month (and are now at +2.9% y/​y). Clearly, there were price pressures outside of tax effects on the month. 

Looking ahead, the path looks a fair bit complication for the Bank of Canada. Indeed, price pressures will probably remain sticky in the months ahead due to:

  • Retaliatory tariffs
  • The indirect costs of the trade war for businesses
  • Any residual impact from the imposition of GST/HST

Offsetting those, you have the elimination of the consumer carbon tax.

In any case, there’s more two-way’ risk for upcoming BoC meetings (two way = either a pause, or a cut) than before. The April/​June/​July BoC meetings are roughly pricing in a 36-40% chance of a 25bps cut now.

b.) In an interview with Fox, Trump says that it’d be easier” to work with a Liberal Party leader in Canada though “(he) really doesn’t care.”

5.) Asia-Pacific Notes:

As expected, the Bank of Japan left policy unchanged. The 9-0 vote points to little internal debate as a second consecutive rate hike was probably not in the cards.

The market currently sees the next hike coming in June.