Q2 2025 Market Update
How Retirement Models are Broken and the Remedy
Current Market Backdrop
The S&P 500 is expensive at 23X forward earnings and near record levels. It is this pricey around 7% of the time over the past 40 years. However, profit margin trends are positive – possibly justifying some of the premium. That’s the analyst narrative anyway.
Our view is that the high valuation has more to do with government deficits which are promotive of money supply growth and market price increases.
In our last market update we brought up the idea that there may be a recession if DOGE/the Gov cut spending like it appeared they intended to. Then we all know what happened during budget negotiations. Congress didn’t want to cut. All the same spending continues, with some minor tax cuts. Basically, Congress caved and took up the growth narrative. Cutting is unpopular. With the current influence of our lobbyists and no term limits for Congress, serious budgeting appears off the table until there is a crisis.
Back to the market. This stimulative spending bill is supportive of premium P/Es because ownership of things like risk assets (stock, gold, crypto and real estate) gain importance in the face of never-ending $2T budget deficits (aka money-printing regimes). This will bring me to my next section.
We still expect a turbulent year for the market, but this is the general backdrop.
The Errors in Retirement Models
If the assumptions are wrong, the model just allows you be wrong with confidence.
Retirement withdrawal modeling such as Monte Carlo simulations are based on assumptions. One of the key assumptions is inflation. Type in a retirement scenario into Chat or Grok such as $3m liquid net worth (not counting main residence), age 65, live to 90 and let AI fill in the rest.
AI (retirement modeling) will give you 3% to 4% of your balance in withdrawals for living per year ($90K to $120K in this case). It will assume a 2.5% to 3% inflation rate. For a moderate risk tolerance, it will recommend around 50% of the portfolio for stocks and stock alternatives like MLPs, REITs. The rest is placed in bonds and treasuries.
The Monte Carlo simulation allows for the exact median withdrawal of $105K per year with a 2.5% increase per year. It allows for a 92% chance of not running out of money.
Now let’s get realistic.
Run the same model with a 5% inflation rate (See table below for 5% inflation assumption) and the probably of not running out of money drops to 64%. “With a 5% annual inflation rate:
• The probability of not running out of money drops to 64.1%. • The median ending balance falls to around $1.08 million.”
A 60/40 allocation with 5% inflation increases the likelihood of not running out of money to 70% according to the Monte Carlo model. A 70/30, or 80/20 allocation does not increase the success rate probability. It remains at 70%. Therefore, the 60/40 allocation is the best split in this scenario when assuming 5% inflation and all other internal assumptions of the Monte Carlo model the same.
But hang on, let’s check the less-known return assumptions the Monte Carlo model uses:
Asset Expected Return Stocks 7.0% Bonds 3.5% Gold 3.0% Bitcoin 12.0%
Do those returns look low to you? These assumptions are strikingly unambitious for stocks, gold, and bitcoin if we look at historical 20-year returns. They are obtained from projections of JP Morgan, Blackrock, Vanguard and others. Since we are looking at a 20 to 25-year time horizon for the investments, let’s adjust the return assumptions to actual returns over the past 20 years (listed below), and use a flat 20% return for bitcoin since it likely will not continue to return 49% per year:
Asset Total Return (Approx.) CAGR (Annualized Return) Gold ~+400% ~8.3% Asset Total Return (Approx.) CAGR (Annualized Return) S&P 500 ~+325% ~7.6% (Dividends included) Bonds ~+100% (AGG ETF) ~3.5% Bitcoin ~49% (over long term)
Stocks have returned 10% annually over the past century but we’ll use the 7.6%. Let’s plug in our new assumptions based on the past 20 years. The new recommended allocation for optimization is:
“General Insight Based on Prior Runs Allocation Prob. of Median Ending (S/B/G/BTC) Success Balance 60 / 25 / 10 / 5 ~84% ~$3.9M 65 / 15 / 10 / 10 ~84% ~$5.9M
Likely Optimal Tradeoff (Per AI based on the new assumptions)
65% stocks / 15% bonds / 10% gold / 10% bitcoin
• Offers the best blend of high success rate and high upside
• Leverages bitcoin’s high long-term CAGR and gold’s non-correlation
• Keeps enough bonds to buffer volatility”
A more realistic 5% inflation rate greatly increased our likelihood of running out of money in retirement. By adjusting the return assumptions to the past 20 year returns, adding gold, and bitcoin to the mix, and changing the allocation to a more aggressive 65/15/10/10, our probability of not running out of money goes back up to 84% while leaving a median ending balance upon death of an impressive $5.9m.
Before allowing a financial planner (and accompanying Monte Carlo simulation) to plan your retirement; understand and challenge the assumptions used in order to come up with an effective allocation.
Back to the inflation discussion. Following are Price Increases (provided by Chat and Grok) for various static products or services in the U.S. over 30 or 50 years. This information was nearly impossible to gather prior to AI.
Average Annual
Time Start Price End Price (2025, Total %
Item % Increase
Period (Estimated) Estimated) Increase
(CAGR)
1975-
Big Mac (U.S. metro cities) 2025 $0.75 $6.00 700.00% 4.11%
(50y)
1975-
Electricity Bill (U.S. metro
2025 $22.00 $200.00 809.09% 4.68%
areas)
(50y)
1995-
Family Healthcare
2025 $5,791.00 $37,720.00 551.38% 6.67%
Coverage (U.S.)
(30y)
1995-
Single Healthcare
2025 $2,152.00 $12,298.00 471.28% 6.15%
Coverage (U.S.)
(30y)
1975-
Car Insurance (U.S.) 2025 $200.00 $2,150.00 975.00% 5.03%
(50y)
1995-
Property Tax (Atlanta, 30y) 2025 $1,200.00 $5,000.00 316.67% 4.92%
(30y)
1975-
Property Tax (Atlanta, 50y) 2025 $400.00 $8,000.00 1900.00% 6.40%
(50y)
Average Annual
Time Start Price End Price (2025, Total %
Item % Increase
Period (Estimated) Estimated) Increase
(CAGR)
1975-
Men’s Haircut (U.S. metro
2025 $2.50 $45.00 1700.00% 5.92%
areas)
(50y)
1975-
2-Bed, 2-Bath Apartment
2025 $175.00 $2,000.00 1042.86% 5.13%
(Atlanta)
(50y)
1975-
2-Bed, 2-Bath Apartment
2025 $192.00 $2,000.00 941.67% 4.82%
(U.S. metro areas)
(50y)
1975-
Can of Baked Beans (U.S.
2025 $0.35 $3.50 900.00% 4.71%
metro areas)
(50y)
1975-
Pound of Raw Cane Sugar
2025 $0.50 $2.50 400.00% 3.32%
(U.S. metro areas)
(50y)
1975-
Dry Cleaning a Pair of
2025 $1.50 $8.00 433.33% 3.43%
Pants (U.S. metro areas)
(50y)
Maid Cleaning Service 1975-
(U.S. metro areas, 2-bed, 2025 $20.00 $175.00 775.00% 4.55%
2-bath) (50y)
1975-
Annual Doctor’s Office
2025 $20.00 $200.00 900.00% 4.71%
Visit (U.S. metro areas)
(50y)
Average: 4.97%
Final Thoughts
It appeared ever so promising that a platform component of reducing spending and the deficit was a priority for the current administration. But that willpower dissipated like confetti at a parade – bright, brief, and forgotten. If a majority Republican Congress and President cave to lobbyists and special interests, and to every group appealing for spending, we think the continuation of reckless fiscal policy is nailed down. This means of course, ‘inflating away the debt,” as the quote goes.
What it means for future and current retirees expecting a long retirement is a needed adjustment in investment in risk assets to a higher allocation. We see no reason to believe true dollar devaluation will be anything less than 4% to 5% annually. For all income and wealth levels, that is an incredibly expensive tax.
"We can’t keep running deficits like this forever without consequences. It’s not sustainable, and inflation is one way it shows up." -Warren Buffett
All investments involve risk, including the possible loss of principal.
The comments, opinions and analyses expressed herein are for informational purposes only and should not be considered individual investment advice or recommendations to invest in any security or to adopt any investment strategy. The opinions and analysis are rendered as of publications date and may change without notice. The information provided in this material is not intended as a complete analysis of every material fact regarding any security, country, region, or market.
