Scott Martindale  by Scott Martindale
  President & CEO, Sabrient Systems LLC

As an update to some of the topics I discussed in my lengthy early-January post, I wanted to share an update in advance of the FOMC announcement on Wednesday based on several economic reports that were released last week.

The Fed’s preferred inflation metric, Core Personal Consumption Expenditures (PCE, excluding food & energy) for December came in on Friday at 2.93% YoY and 0.17% MoM. However, I prefer to focus on the most recent trend over the past 3 months. Annualizing the Core PCE price index change over the past 3 months (0.14% in Oct, 0.06% in Nov and 0.17% in Dec) computes a 1.52% annualized inflation rate, as shown in the chart below.

So, it appears to me that inflation today is likely below the 2% target such that the Fed can begin normalizing fed funds rate toward the “neutral rate” (neither contractionary nor expansionary), which I believe ultimately will be around 3.0% nominal (i.e., 2% target inflation plus 1.0% r-star) versus 5.25–5.50% today. Moreover, I think the 10-year Treasury will settle at about 4.0–4.50%, assuming we don’t continue flood the Treasury market with new issuances to fund fiscal boondoggles and rising debt service (we’ll get a clue on Wednesday with the Treasury Refunding announcement). The rate levels I am suggesting seem appropriate so that borrowers can handle the debt burden while fixed income investors can receive a reasonable real yield (i.e., above the inflation rate) so they don’t have to take on undue risk to achieve meaningful income.

The Fed believes the composition of PCE more accurately reflects current impacts on consumers than does CPI. This is because it more quickly adapts to consumer choices through its weighting adjustments to individual items (e.g., shifts from pricier brands to discount brands). Also, while CPI narrowly considers only urban expenditures, PCE considers both urban and rural consumers as well as third-party purchases on behalf of a consumer, such as healthcare insurers buying prescription drugs. Furthermore, items are weighted differently—for example, shelter is the largest component of CPI at 32.9% but only 15.9% of PCE, and healthcare is the largest component of PCE at 16.8% but only 7.0% of CPI. So, while Core PCE shows a 1.52% 3-month annualized inflation rate, Core CPI is 3.33%.

So, let's talk more about shelter cost. I have often discussed the lag in shelter cost metrics distorting both PCE and CPI, particularly as new leases gradually roll over throughout the course of a year while existing lease rates persist. So, let me introduce another metric published by the BLS that provides more current insights into the trend in shelter cost, namely the New Tenant Rent Index (NTR).

The NTR data peaked in Q2 2022 while CPI Shelter didn’t peak until Q2 2023. And NTR has fallen precipitously since then, showing a substantial -8.75% quarter-over-quarter decline in Q4 2023 versus Q3 2023. But because such QoQ comparisons can be quite volatile with this metric, I’m not going to annualize that number. Instead, let’s stick with the year-over-year or 4-quarter comparison, which shows a more modest (but still significant) decline of -4.74% versus Q4 2022. Although CPI Shelter index is still elevated, it is also falling (as shown in the chart), now showing a YoY rate of 6.15% for December.

Inflation metrics

Furthermore, the BLS also publishes an All Tenant Regressed Rent Index (ATRR), which is not restricted only to new leases, so it moves more slowly and with less volatility. ATRR also has been in a steady decline since peaking in Q4 2022 at 7.84%, and it has been steadily falling over the past 4 quarters to its latest Q4 2023 reading of 5.27% YoY, which again reflects rapidly falling rental prices and is in-line with CPI Shelter.

This suggests to me that falling shelter costs will soon be more impactful to PCE and CPI readings. The FOMC is surely aware of this.

As for other economic reports last week, we saw the BEA’s advance (first) estimate of Q4 GDP growth surprised to the upside at a 3.3% annual rate, largely driven by personal consumption. Also, the December reading for M2SL money supply shows it has stayed basically flat since last March, while velocity of money (M2V) continues to ramp up. This suggests that more transactions are occurring in the economy for each dollar in circulation, which has offset the negative impact of stagnant money supply, thus supporting GDP, corporate earnings, and stock prices—although lack of M2 growth creates other strains on liquidity. I discuss this further in today’s post below.

Click here to continue reading my full commentary. And please feel free to share my full post with your friends, colleagues, and clients! You also can sign up for email delivery of this periodic newsletter at Sabrient.com.

  Scott Martindaleby Scott Martindale
  President & CEO, Sabrient Systems LLC

To be sure, 2023 was another eventful year (they just keep coming at us, don’t they?), ranging from escalating hot wars to a regional banking crisis, rising interest rates, falling inflation, a dire migration crisis, and an AI-driven frenzy in the so-called “Magnificent Seven” (MAG7) corporate titans— Meta Platforms (META, ne: FB), Apple (AAPL), Nvidia (NVDA), Alphabet (GOOGL), Microsoft (MSFT), Amazon (AMZN), and Tesla (TSLA), aka “FANGMAT,” as I used to call them—which as a group contributed roughly 60% to the S&P 500’s +26.2% gain in 2023. Their hyper-growth means that they now make up roughly 30% of the index. Nvidia (NVDA), whose semiconductors have become essential for AI applications, was the best performer for the full year at +239%.

Small caps finally found some life late in the year, with the Russell 2000 small cap index essentially keeping up with the S&P 500 starting in May and significantly outperforming in December. Bonds also made a big comeback late in the year on Fed-pivot optimism, which allowed the traditional 60/40 stock/bond allocation portfolio to enjoy a healthy return, which I’m sure made a lot of investors and their advisors happy given that 60/40 had been almost left for dead. The CBOE Volatility Index (VIX) has been below 20 for virtually the entirety of 2023 and as low as 11.81 in December, closing the year at 12.45. Also, as a breadth indicator, the percentage of stocks that finished the year above their 200-day moving average hit 75%, which is bullish.

Nevertheless, the Russell 2000 (+16.8%) and the equal-weight version of the S&P 500 (+13.7%) were up much less for the full year than the cap-weighted S&P 500 (+26.2%) and Nasdaq 100 (+54.9%). In fact, 72% of the stocks in the S&P 500 underperformed the overall index for the full year, illustrating that despite the improvement in breadth during the second half of the year, it could not overcome the huge outperformance of a small cohort of dominant companies. This suggests that either the market is set up for a fall in 2024 (as those dominant companies sell off) …or we’ll get a continued broadening into other high-quality companies, including mid- and small caps. I think it will be the latter—but not without some volatility and a significant pullback. Indeed, despite signaling investor confidence and complacency by remaining low for a long stretch, the VIX appears to be ripe for a spike in volatility. I think we could see a significant market correction during H1 (perhaps to as low as 4,500 on the S&P 500) even if, as I expect, real GDP growth slows but remains positive and disinflationary trends continue, supporting real wage growth and real yields—before seeing an H2 rally into (and hopefully following) the November election. And don’t forget there’s a potential tsunami of cash from the $6 trillion held in money market funds, as interest rates fall, much of it may well find its way into stocks.

Not surprisingly, last year ended with some tax-loss harvesting (selling of big losers), and then the new year began last week with some tax-gain harvesting—i.e., selling of big winners to defer tax liability on capital gains into 2024. There also has been some notable rotation of capital last week into 2023’s worst performers that still display strong earnings growth potential and solid prospects for a rebound this year, such as those in the Healthcare, Utilities, and Consumer Staples sectors. Homebuilders remain near all-time highs and should continue to find a tailwind as a more dovish Fed means lower mortgage rates and a possible housing boom. Energy might be interesting as well, particularly LPG shipping (a big winner last year) due to its growing demand in Europe and Asia.

As I discussed in my December commentary, I also like the prospects for longer-duration bonds, commodities, oil, gold, and uranium miner stocks this year, as well as physical gold, silver, and cryptocurrency as stores of value in an uncertain macro climate. Also, while Chinese stocks are near 4-year lows, many other international markets are near multi-year highs (including Europe and Japan), particularly as central banks take a more accommodative stance. Indeed, Sabrient’s SectorCast ETF rankings show high scores for some international-focused ETFs (as discussed later in this post).

While stocks rallied in 2023 (and bonds made a late-year comeback) mainly due to speculation on a Fed pivot toward lower interest rates (which supports valuations), for 2024 investors will want to see more in the way of actual earnings growth and other positive developments for the economy. I expect something of a “normalization” away from extreme valuation differentials and continued improvement in market breadth, whether it’s outperformance by last year’s laggards or a stagnation/pullback among last year’s biggest winners (especially if there are fewer rate cuts than anticipated)—or perhaps a bit of both. Notably, the S&P 500 historically has risen 20 of the last 24 election years (83%); however, a recent Investopedia poll shows that the November election is the biggest worry among investors right now, so it’s possible all the chaos, wailing and gnashing of teeth about Trump’s candidacy will make this election year unique with respect to stocks.

Regardless, I continue to believe that investors will be better served this year by active strategies that can identify and exploit performance dispersion among stocks across the capitalization spectrum—particularly smaller caps and the underappreciated, high-quality/low-valuation growers. Small caps tend to carry debt and be more sensitive to interest rates, so they have the potential to outperform when interest rates fall, but you should focus on stocks with an all-weather product line, a robust growth forecast, a solid balance sheet, and customer loyalty, which makes them more likely to withstand market volatility—which may well include those must-have, AI-oriented Tech stocks. Much like the impact of the Internet in the 1990s, AI/ML, blockchain/distributed ledger technologies (DLTs), and quantum computing appear to be the “it” technologies of the 2020’s that make productivity and efficiency soar. However, as I discuss in today’s post, the power requirements will be immense and rise exponentially. So, perhaps this will add urgency to what might become the technology of the 2030’s—i.e., nuclear fusion.

On that note, let me remind you that Sabrient’s actively selected portfolios include the Baker’s Dozen (a concentrated 13-stock portfolio offering the potential for significant outperformance), Small Cap Growth (an alpha-seeking alternative to a passive index like the Russell 2000), and Dividend (a growth plus income strategy paying a 4.5% current yield).

By the way, several revealing economic reports were released last week, which I discuss in today’s post. One was the December reading on the underappreciated New York Federal Reserve Global Supply Chain Pressure Index (GSCPI), which has fallen precipitously from it pandemic-era high and now is fluctuating around the zero line. This historically suggests falling inflation readings ahead. As for the persistently inverted yield curve, I continue to believe it has more to do with the unprecedented supply chain shocks coupled with massive fiscal and monetary stimulus to maintain demand and the resulting surge in inflation, which as observed by Alpine Macro, “makes the inversion more reflective of different inflation expectations than a signal for an impending recession.”

Also, although M2 money supply fell -4.6% from its all-time high in July 2022 until its low in April 2023, it has essentially flatlined since then and in fact has been largely offset to a great extent by an increase in the velocity of money supply. Also, we have a robust jobs market that has slowed but is far from faltering. And then there is the yield curve inversion that has been gradually flattening from a low of about -108 bps last July to -35 bps today.

I discuss all of this in greater detail in today’s post, including several illustrative tables and charts. I also discuss Sabrient’s latest fundamentals based SectorCast quantitative rankings of the ten U.S. business sectors (which is topped by Technology), current positioning of our sector rotation model (which turned bullish in early November and remains so), and some actionable ETF trading ideas.

Overall, I expect inflation will resume its decline, even with positive GDP growth, particularly given stagnant money supply growth, mending and diversifying supply chains (encompassing manufacturing, transportation, logistics, energy, and labor), falling or stabilizing home sale prices and new leases, slowing wage inflation, slower consumer spending on both goods and services, and a strong deflationary impulse from China due to its economic malaise and “dumping” of consumer goods to shore up its manufacturing (US imports from China were down 25% in 2023 vs. 2022). This eventually will give the Fed (and indeed, other central banks) license to begin cutting rates—likely by mid-year, both to head off renewed crises in banking and housing and to mitigate growing strains on highly leveraged businesses, consumers, government, and trading partners. Current CBOE fed funds futures suggest a 98% chance of at least 100 bps in rate cuts by year end (target rate of 4.25-4.50%), and 54% chance of at least 150 bps.

Click here to continue reading my full commentary … or if you prefer, here is a link to this post in printable PDF format (as some of my readers have requested). And please feel free to share my full post with your friends, colleagues, and clients! You also can sign up for email delivery of this periodic newsletter at Sabrient.com