Broker Check

Quarterly Newsletter

What Would You Have To Believe?                                                                                          Join our Quarterly Newsletter          April 2024                                                                                             

Mark R. Hoffman

CEO, Principal

Over the years and across a wide range of topics, I have often found myself going through a mental exercise that I call “What Would You Have To Believe?”  It’s a simple concept, but at least for me, it can help me think through alternative scenarios to either prove or disprove a hypothesis. 

Recently, I have found myself using that exercise when I think about U.S. stock market returns.  By any measure, the U.S. stock market has been on a tear.  The S&P 500 rose over 20% in 2023, driven by a ridiculous 76% gain in the Magnificent 7 (Amazon, Alphabet, Microsoft, Apple, Tesla, Meta and Nvidia).  2024 started off with more of the same until the last 3-4 weeks have slowed the pace of the return. 

This bull run cannot last forever.  We will have a correction.  We always do.  But I have learned that bull runs can last a lot longer and go a lot higher than many would expect.  So…what would I have to believe to think that the bull run can continue from here? And how realistic are those scenarios?  Simple math tells us that to have the stock market go up from any level, you would need any or all of the following: dividend growth, multiple expansion or earnings growth. 

Dividends.  What would I have to believe to have dividend growth drive the S&P 500 higher from here?  I would have to believe that the S&P 500 dividend yields would buck the 60-year trend that has seen dividend yields drop from 3-4%/year to less than 2%/year.  In fact, the only times yield has gotten close to or over 3%/year over the last 30-years have been following the financial crisis and COVID – both of which resulted in stock prices cratering which is the exact opposite of stock prices continuing to climb.  Short answer here is that I don’t believe it can be dividends. 


Multiple expansion.  What would I have to believe to have multiple expansion drive the S&P 500 higher from here?  Looking at the long-term chart of trailing twelve month P/E multiples, we currently sit at 27.6x, which is a lofty multiple.  For multiple expansion to continue the current bull run, I would have to believe that the new trend line we should look at begins around 1980 and that everything before that is just noise.  Since 1980 (with an extreme amount of volatility), multiples have expanded from a little less than 10x to where they are today – in the high 20s.  Do I believe that the new average is somewhere closer to 20x than the 15x average we saw for the 100 years leading up to 1980?  Yes I do.  But even if the new average is closer to 20x, we are 30-40% higher than that right now.  So do I believe multiple expansion is what is going to keep this run going?  I do not.


Earnings growth.  That leaves earnings growth.  What would I have to believe there?  Company earnings are a factor of the number of employees times the productivity per employee.  Do I believe that we can grow our way into the new multiples by increasing the number of workers in America?  Short answer is no.  With baby boomers retiring, the unemployment level near all time lows, the birth rate in the U.S. declining and our country’s current stance on immigration, we will not solve our earnings issue with more workers.  So it would have to be productivity.

Productivity of the U.S. workforce has been below 2%/year since 2007. 


Is there anything that would cause you to believe that productivity growth could once again get back to 2%/year for the first time since the development of the internet?  In this case, I think there might be.  Artificial intelligence (AI). 

In the simplest of terms, AI is technology that allows computers to learn from experience and simulate human intelligence to solve problems.  AI is the next true game changer.   Amazon CEO Andy Jassy states in his latest letter to Amazon shareholders: “…AI may be the largest technology transformation since the cloud (which itself, is still in the early stages), and perhaps since the Internet...The amount of societal and business benefit from the solutions that will be possible will astound us all.”  You might expect Amazon’s CEO to say something like that given that this is his business.  But if you think he’s alone, read any of the hundreds of projections that are out there on what AI might be able to accomplish. 

I’ll save you the search.  Here are a handful of the estimated productivity benefits (data/estimates from a variety of sources including McKinsey, PWC, Deloitte, etc.):

  • Healthcare: 40-60% reduction in drug development time and cost; 30-50% reduction in diagnostic imaging interpretation time with 20-30% improvement in accuracy; 20-30% reduction in paperwork and administrative burden
  • Manufacturing: 30-50% reduction in maintenance cost and 20-30% reduction in downtime from predictive maintenance; 25-45% improvement in inventory optimization; 15-25% reduction in defects
  • Finance: 30-50% improvement in fraud detection rates with 20-35% reduction in false positives; 20-30% reduction in trading errors
  • Retail: 25-45% reduction in out-of-stocks from inventory optimization; 30-50% improvement in demand forecasting
  • Transportation: 25-45% reduction in maintenance cost and 20-35% improvement in downtime from predictive maintenance; 15-30% reduction in fuel cost from route optimization
  • Agriculture: 20-40% improvement in crop yields from precision farming; 20-35% reduction in water usage and 25-45% reduction in crop loss from pests and disease from crop monitoring

The productivity improvement potential is outrageous.  I don’t know about you, but I certainly wasn’t thinking about water usage from crop monitoring when I was considering potential AI applications.  Amazon’s Jassy may be right.  Even if you are a skeptic, cut all these estimates in half and prolong the time it takes to get these results, it’s easy to see how these types of applications could take annual productivity change in the chart above back above 2% - perhaps well above 2%.  The impact on global GDP would be measured in the trillions. 

Artificial intelligence is not going to stop stock markets from correcting.  Markets always race too far too fast and set themselves up for a down period.  Exogenous shocks happen.  But of all the possible drivers of a continued bull market, I have to believe that AI has the most potential to dramatically improve earnings across a very wide range of businesses, allowing those companies to backfill the lofty P/E multiples that have driven the markets to the current levels.  It will take some time.  But I don’t think I’d bet against it.  

Mark is a co-founder of Lanier Asset Management and serves as its Chief Executive Officer. Prior to founding Lanier, he was a partner at The Boston Consulting Group. Mark is an honors graduate of The University of North Carolina at Chapel Hill with a BA in Economics, and holds an MBA from The Harvard Business School.