Are Sub-100k NFPs the New Normal?

Summary

• Recent NFP prints and pre-pandemic CBO projections suggest sub-100k NFPs could be the new normal now the immigration surge is over.
• If so, a large negative surprise at the 6 December NFP release is likely, which would rekindle recessions fears and expectations of Fed cuts.
• I expect the Fed to respond with a cut at the December meeting.
• Longer term, because the low NFP would come with low unemployment, sticky wages and inflation, the Fed would be unable to cut further.

Market Implications

• No Fed cut in 2025, though I still expect a 25bp cut in December.

Sub-100k NFPs Likely New Normal

In this note, I discuss the risks and implications of a new normal: much lower than consensus NFPs. I assume that, over the next year or so, deportations of illegal immigrants are too limited to impact NFPs.

NFPs have averaged 207,000 in H1 and 148,000 in Q3. October NFPs were 12,000. Also, there were 44,000 workers on strike and therefore not included in NFPs. I found little evidence of hurricane impact. Even if the total impact of strikes and hurricane was higher than the 44,000 strikers, the underlying NFP print was likely well below 100,000.

The lower NFPs likely reflects far fewer migrants after the administration closed the border in June. The CBO estimates net immigration was 2.6mn in 2022 and 3.3mn in 2023, and contributed about 80% of total population growth in both years. This compares with 0.9mn annual average immigration during 2010-19 (Table 1). Most of the recent increase was in the ‘other immigrant category’, i.e., illegal migrants and asylum parolees (Table 1).

Before the June border closure, the CBO forecast immigration would increase by 3.3mn in 2024 and 2.6mn in 2025, leading to average NFPs of 195,000 and 130,000.

The Q3 NFP slowdown likely largely reflects the immigration surge ending. Employment in sectors traditionally more reliant on illegal migrants slowed more than average (Table 2). These sectors include agriculture, construction, low-skill services. The slowdown has been greatest in agricultural employment, taxi driving, personal services (that includes nannies and housekeepers), and waste management.

Furthermore, CBO projections published before the pandemic suggest NFPs well below 100,000 could be the new normal. In January 2020, the CBO projected NFP at 50,000 on average in 2025-30 due to unfavourable demographics, including baby-boomer retirements (Table 1). The CBO then stressed that in 2018, net immigration already accounted for 48% of population growth. By 2050, the CBO expected net immigration to account for all of population growth.

Since the January 2020 CBO projections, mortality has increased, and fertility decreased. In its 2024 projections, the CBO expects net immigration to account for all population growth by 2040.

The first test of my view is likely to come at the 6 December release of the November employment report.

Recession Expectations to Rekindle

Current consensus for the November NFPs is 175,000, based on three submissions. If consensus remains around these levels and we get a sub-100k number as I expect, expectations of recessions are likely to surge. Long-term yields are likely to fall, and expectations of FFR cuts are likely to rekindle, i.e., the curve is likely to steepen. Market reaction could be strong as hedge funds are short across the curve. Equities would likely sell off and the dollar weaken.

In addition, if my view is correct, we are likely to see the economic surprise index, currently close to the top of its recent range, turn around (Chart 1). This is because, in addition to lower NFPs, lower immigration is likely to cause slower consumption and therefore GDP growth. Indeed, the Atlanta Fed nowcast, my preferred macro data aggregator, has remained below Q3, so far (Chart 2; the nowcasts are based on the same data releases every quarter).

In addition, NFP forecasts are likely to remain biased.

NFP Forecast Bias to Flip

NFPs have exhibited a striking and persistent forecast bias over the past two years (Chart 3). Pre-pandemic, forecast errors tended to revert to zero. By contrast, since 2022 Q2, forecasts have consistently underestimated payrolls (Chart 3).

I think the most likely explanation for the bias is a misinterpretation of the rise in unemployment: consensus likely saw it as a sign of weaker demand for labour.

In reality, it reflected the immigration surge – it was a sign of stronger labour supply. Low-skill migrants took longer to find jobs, lowering the job-finding rate and raising unemployment duration and uninsured unemployment. All the increase in unemployment since 2023 has been in uninsured unemployment (Chart 4). Likely, the uninsured unemployed were mostly undocumented migrants who either had not contributed long enough to qualify for UI or could not draw benefits because of their legal status.

There is a good chance that the forecast bias will persist but flip: if my view is correct, low NFPs will not see an increase in unemployment. Unemployment could actually fall as we saw in August-September. If the consensus ignores the slower labour supply growth, it is likely to expect low unemployment to translate into high NFP.

Fed Cuts Limited by Negative Labour Supply Shock

The Fed would likely cut in response to a sub-100,000 NFP print on 6 December but subsequently stay on hold even if the sub-100,000 NFPs continue. (6 December happens to be the last day before the start of the pre-meeting blackout – so if a cut is not priced in, it could get announced through the proverbial Timiraos dispatch.)

Until now, the Fed inflation and employment objectives have not conflicted. I do not expect the next inflation print would prevent the Fed from responding to a weak NFP print.

This is because the Fed is comfortable with inflation remaining at current levels for some time. In his last speech on 14 November, Powell stated that the Fed was ‘watching carefully’ to ensure core goods and super core inflation remained in their ‘recent ranges’ and that housing inflation eventually normalized.

The next inflation print (the last one before the FOMC) would have to lift core PCE well above the current range around 2.75% to preclude a cut. That is, it would have to be well above 25bp MoM. I do not think it likely as inflation moves slowly and past 3m and 6m averages have been 22bp and 19bp.

The Fed is therefore likely to respond to a sub-100k NFP with a cut. This is because in the short run, it is difficult to distinguish between a demand-driven and supply-driven low print. In either case, slower NFP growth is likely to come with slower GDP growth.

Labour demand-driven NFP weakness would come with other signs of labour market weakness, mainly higher unemployment and lower wages. By contrast, labour supply (lower immigration)-driven NFP weakness would see only limited changes in unemployment and wage growth. The labour market could actually tighten. But the data is noisy, and these patterns could take a few months to appear clearly.

In addition, and not unreasonably, the FOMC doves would argue that a sub-100k NFP tilts the balance of risks towards greater employment downside.

Finally, the Taylor rule, which has reliably indicated Fed policy in this cycle, shows the Fed still has room to cut (Chart 5).

I do not expect the Fed to be able to cut beyond December. If my view that NFP weaknesses are driven by low immigration is correct, unemployment, wages and, beyond the labour market, the output gap will change little and could even signal greater imbalance between aggregate demand and aggregate supply. Inflation would therefore be likely to remain stuck in the current range, which would prevent the Fed from cutting in 2025 (Chart 6).

The Fed’s inability to ease would see the short end of the curve sell off and the curve flatten. Equity markets would be hit by the double whammy of lower growth and limited Fed easing. This would be a form of payback for the positive supply shock represented by the 2002-24 immigration surge. This shock is largely what allowed disinflation with above trend growth.

Market Consequences

I think markets underprice a December cut. They show only a 50% chance of a December cut when my conviction is closer to 75%. By contrast, markets overprice 2025 cuts as they see 60bp and I expect none.

.

(The commentary contained in the above article does not constitute an offer or a solicitation, or a recommendation to implement or liquidate an investment or to carry out any other transaction. It should not be used as a basis for any investment decision or other decision. Any investment decision should be based on appropriate professional advice specific to your needs.)