Productivity Growth Falls & Unit Labor Cost Growth Rises

In Q3, productivity growth fell and unit labor costs rose. That’s good for workers and bad for profit margins. We are now in the longest margin contraction in post war history as NIPA profit margins have fallen for 4.5 years. It’s unusual for margins to fall this far for this long without a recession. It hasn’t happened to this extent since at least 1960. Despite the decline in margins due in part to high wage growth, there haven’t been layoffs as jobless claims are very low.

Specifically, in Q2 quarterly annualized productivity growth was revised higher from 2.3% to 2.5%. Q3 was the opposite as growth fell to -0.3% which missed estimates for 1% and the low end of the consensus range which was 0.1%. That was the first decline in productivity in 15 quarters. Hours worked grew quicker than output which weighed on productivity growth.

Quarterly annualized unit labor cost growth in Q2 was revised from 2.6% to 2.4%. Q3 was the opposite again as growth was 3.6% which beat the consensus of 2.2% and the high end of the consensus range which is 2.9%. Compensation per hour growth was 3.3%. 

In manufacturing, productivity growth was 0.1% which increased from -2.4%. Output growth was 1.1% and hours worked growth was 1.3%. Compensation growth was 3.5% and unit labor cost growth was 3.6%. Yearly productivity growth fell from 1.8% to 1.4% in Q3 which is more of the same for this expansion. Growth has been very low in the past decade. As you can see from the chart below, productivity growth has averaged 1.32% this cycle which is below the long term average of 2.15%.

(Click on image to enlarge)

Strong Purchase Applications Growth

In the week of November 1st, the MBA applications report contracted on a weekly basis, but had solid yearly purchase applications growth. Composite index was down 0.1% weekly after rising 0.6% in the prior week. Purchase index fell 3% after rising 2% in the prior week. The good news is yearly growth is still a solid 7%. Refinance index was up 2% weekly after falling 1% in the prior week. 

In the week of October 31st, which is the week of this report, the average 30-year fixed-rate increased 3 basis points to 3.78%. A recent trough was 3.49% in September. In the week of November 7th, the rate fell 9 basis points to 3.69%. That decline won’t continue though as long rates have recently spiked.

On the one hand, the chart below shows the comps are easy and will stay easy for the next few weeks. On the other hand, it’s good to see growth in the face of rising rates. With the big spike in long rates on Thursday, demand for housing could come under pressure. 

Rates are still much lower than they were last year and all housing stats will face easy comps this fall, but on a sequential basis we could see weakening demand. Higher rates are problematic at a time when consumers view housing as unaffordable.

(Click on image to enlarge)

Fannie Mae Home Purchase Sentiment index fell 2.7 points in October to 88.8. It’s up 3.1 points from last year. Comps are easy. Net percentage of Americans who think now is a good time to buy a house fell 7 points to 21%. And, net percentage who say now is a good time to sell a house fell 3 points to 41%. 20% more say now is a good time to sell than buy. 

Net percentage of Americans saying home prices will increase fell 2 points to 27%. This percentage has been falling since June. Home price growth has been weak but has recently been stabilizing. Growth can easily plummet further if the increase in long rates continues. Many investors predict the 10-year yield will get to 2%. That means mortgage rates will rise a bit more.

Net percentage of Americans who think mortgage rates will rise in the next year fell 2 points to -25%. It’s pretty obvious that mortgage rates won’t rise as they are near their record low. Consumers are smart enough to know this. 

Good news is the net percentage of Americans who aren’t concerned about losing their job increased 3 points to 72%. This goes against the modest negativity on future expectations of the labor market in the Conference Board report. Finally, the net percentage of Americans who say their household income is significantly higher than it was 1 year ago fell 5 points to 16%.

Jobless Cuts Fall: They Are Still Very Low

Jobless claims have barely been moving in the past few weeks. The 12-week range of the 4-week average is only 5,000. The only 2 other times the range was tighter were in 1973 and 2017. To be clear, the range is likely to be tighter when claims are low. 

In general, volatility in claims spikes in recessions as you can see from the chart below. Specifically, jobless claims in the week of November 12th fell from 219,000 to 211,000. That’s a very low reading as it was 4,000 below the consensus and is only 18,000 above the cycle low in April.

(Click on image to enlarge)

Personally, I see almost no chance of the number of claims reaching a new cycle low unless a highly unusual one-time event occurs. However, I also don’t see the number of claims spiking. Especially since there likely won't be a recession occurring in the next 18 months. To be clear, the 4-week average rose 250 to 215,250. Continuing claims rose 2,000 to 1.692 million and the 4-week average was unchanged at 1.687 million.

Conclusion

Productivity growth was low in Q3. That’s nothing new as it has been low the whole cycle. Unit labor cost growth was high which lowered profit margins which have been declining for 4.5 years. MBA applications report was solid even though rates are starting to move up. I’m interested to see how the rise in rates on Thursday impacts housing demand. Jobless claims were low again. There has been almost 0 volatility in this report.