Roth vs traditional is a simple mathematical calculation. The problem is key inputs to that calculation are not known. For example, what will be their tax rate during retirement? What is the annualized gain in the 401k for how many years?
If you assume the same (or higher) tax rate in retirement as present and getting a significant return on 401k investments, then it should be an easy decision – Roth. However, if instead you assume tax rate drops from 24% tax rate while working to 15% during retirement and 7% annualized return, then traditional has higher after tax total at retirement. With 20% tax rate while working and 15% during retirement, then it depends how far you are from retirement – Roth has higher total if 33+ years from retirement, traditional if <33 years from retirement.
I think of saving towards downpayment vs 401k investments as a completely separate decision. From a long term financial perspective, maxing out the 401k is most likely going to be a better option. A tax advantaged account earning total market type return is most likely going to have a better long return than the much lower historically averaged home value increase - property tax, - home maintenance costs - opportunity cost from not as open to higher paying out of area job or working fully remote in LCOL area - … Rent savings are rarely enough to overcome this.
The key benefits to owning a home are often personal ones, rather than financial ones. Owning a home often improves quality of life and those quality of life benefits may be worth the relative financial loss Whether it makes sense to prioritize saving for a downpayment or not depends on the specific situation, including the specific personal values.
He had been splitting between 401K and Roth 401K, but when I started to look at how soon he should reach the 32% marginal rate, it seemed like fully-funding the Roth 401K at the 24% marginal rate didn’t seem like a bad gamble.
He will also fund a non-deductible IRA that he can then convert to Roth IRA.
And, yes, I did factor in his 5% state tax rate. It is hard to imagine that he will live in a state w/o income taxes, but retirement is a LONG way off.
I also imagine that we will see higher Fed tax rates in the future, but that is impossible to predict.
Agree with you about this being impossible to calculate w/o those key inputs.
When discussing this with my son, I just assumed $22K contribution/year for 40 years and compounding growth rates ranging from 4% to 7%. The 401K contribution limit will most likely increase during this 40 year period , but assuming it remains constant, the account balance at 65 could range from more than $2 million to more $4.6 million. RMD could be nearly $200K at age 75.
I am guessing that the RMD table will change as the required RMD age ticks up through 2033.
I also assumed no employer match, just to keep things simple.
Again, so many assumptions. I am not firm in my belief of fully funding the Roth 401K–just laying out the inputs I considered and looking for discussion and/or someone to poke holes in my theorizing!
He is 24 with many years of work ahead of him. At the moment, he does not have any use for the money, but would he be better off using the tax savings of 401K contributions and investing in a brokerage account and paying a 15% LTCG rate? He has no interest in actively managing investments so I don’t think he would be able to take advantage of capital gains.
I decided to leave excess 529 funds for unborn grandchildren. I believe that my son can fund a non-deductible IRA and then convert it via Backdoor Roth. I may be using the wrong terms here.
He is able to save in addition to the employer Roth 401K & non-deductible IRA. Employer does not offer a Mega Backdoor Roth. No plans to buy anytime soon.
don’t forget that tax brackets will be changing in '26 absent any Congressional action. In addition, the SALT limitation goes away but the Minimum Tax returns for a lot of folks.
Not once you take into account the leverage inherent in home ownership. If your home price triples over a couple of decades and you only put down 20%, then your return is 1000%, which is highly unlikely to be realized through stock market investments. Add to that the more favorable tax treatment (long term capital gains with $250-$500K exemption) compared to tax-deferred retirement accounts where all withdrawals are taxed as current income.
Purely on the post-tax result in retirement (assuming identical investments and returns), Roth is better if current income tax rate is lower than in retirement, while traditional is better if current income tax rate is higher than in retirement. However, tax rate in retirement is not known now.
But there are other factors, such as differences in rules between Roth and traditional (e.g. required minimum distribution rules). Also, someone who is trying to put in the maximum amount will find that Roth effectively gives a higher maximum amount if the nominal contribution limit is the same (convert both to either pre-tax or post-tax dollars to compare).
Indeed it does, but we’re not unusual in having made vastly more on our house than on any other investment. Also worth noting that an era of high inflation makes the leverage more likely to work in your favor.
One can make leveraged investments in the market as well. However, there is a reason why this is generally not recommended. Leveraged investments are not always winners, and one has to pay interest for the leverage. It’s not free., like the apparent free mortgage in your example. You may win big or lose big, with leverage.
That said, I’ll go through the specific numbers you listed, which have some important missing components in the return calculation, as well as the stock market investment comparison.
The historical average home equity gain per year is ~4% in the US. If I use 4.5%, I get 2.4M after 20 years. $2.4M - $1.0M = $1.4M gain… After capital gains tax above $250k/$500k limit, realtor/sale expenses, and expenses to get property for sale; the net return might drop from $1.4 to approximately $1M.
Suppose the buyer put 20% down, with an $800k mortgage at the current average mortgage rate of 7%. That works out to $940k in interest payments in the first 20 years. Assuming a 1% property tax, this works out to another $340k in property tax . The total between the 2 is $1.3M. I’ll use $1M to reflect the tax benefits. That drops the net gain from $1.0 to $0.
If we assume another 1-2% in home insurance, maintenance, and other new home expenses, that might be another $500k over 20 years, pushing the total to a $500k loss.
One also needs to consider amount saved on rent, which is highly variable depending on what they’d rent in place of the home. I’ll use $30k per year in this example, which works out to $600k, pushing the overall return of $600k gain over 20 years for purchasing a home, prior to inflation.
The S&P 500 averages >10% per year, but I’ll use 8% per year, as 401k investments may become more conservative over the 20 year period than a full market index. The initial $200k investment after 20 years, is expected to be worth an average of $200k*1.08^20 = $930k, for a $730k gain. If the buyer also invests the other extra home expenses mentioned above in the market each year, then the gain increases to well over $1M… more if there is an employer match on the 401k. If the 401k is Roth, there are no taxes on this gain. If there are taxes on the gain, those taxes can be spread out over multiple years, rather than requiring a lump sum, like with a home sale.
One can choose different numbers with different outputs, but my point is that with typical numbers, investing in 401k is financially advantageous to saving for a home downpayment. Even if you choose a low mortgage rate, it is hard to make home downpayment svaings advantageous over 401k savings.
Just to start with this makes no sense, it is a cap rate of 3% on the initial home price and most certainly won’t remain fixed for 20 years.
And you have chosen other unfavorable figures for the comparison: the interest rate is at a 20+ year high (and therefore can be refinanced lower) and the Case-Schiller last 10 year growth is over 7% (and even with the early 2000s bubble about 5% over 20 years). Whereas your assumption of 8% (real?) growth in stock market investments is pretty optimistic at current valuation levels.
Housing may be a bit bubbly at the moment, but if we are shifting to an era of higher inflation, then that is likely to be positive for future (nominal) housing prices, and it’s much less clear that it will be good for the stock market. If the government is inflating away debts, then it is great to have secured non-callable debt like mortgages.
As stated, I was not considering inflation in many of the variables. If you instead express it as 3% of home value or increase of a reasonable percentage per year, the conclusion is the same.
the Case-Schiller last 10 year growth is over 7% (and even with the early 2000s bubble about 5% over 20 years). Whereas your assumption of 8% (real?) growth in stock market investments is pretty optimistic at current valuation levels.
A comparison of Case-Shiller vs S&P 500 over the periods you listed is below, assuming dividend reivestment. In all listed time intervals, the S&P 500 averaged more than 5 percentage points above CS, and the 30 year S&P average, was well above the 8% used in the estimate. In contrast, CS averaged 4-5% over long periods, almost exactly matching the 4.5% rate used in the calculation… only ~2% more than inflation. It’s true that there was a post-COVID real estate bubble, which relates to why CS is so much higher for the past 10 years than any other listed interval You think assuming the COVID-influenced abnormally high rate of real estate gains will continue for 20 years is realistic, but expecting 2% below historical average market gains over a 20 year period is too optimistic?
Case-Shiller vs S&P 500 Annualized (nominal)
Past 10 Years – CS = 6.5%, S&P 500 = 13.5%
Past 20 Years – CS = 4.4%, S&P 500 = 10%
Past 30 Years – CS = 4.6%, S&P 500= 10%
Case-Shiller vs S&P 500 Annualized (inflation adjusted)
Past 10 Years – CS = 4%, S&P 500 = 11%
Past 20 Years – CS = 2%, S&P 500 = 8%
Past 30 Years – CS = 2%, S&P 500 = 7%
And you have chosen other unfavorable figures for the comparison: the interest rate is at a 20+ year high (and therefore can be refinanced lower)
Considering long term treasury bond rates, the market does not expect the fed rate to drop back down to ~0% any time soon… maybe never, so I wouldn’t assume you can just wait for 30-year mortgage rates to drop back down to the historic lows from the last decade, then refinance. If you look back before the Great Recession of 2007-09, 7% mortgage rates were normal, even below historical averages.
That said, as stated in the original post, if you choose a low mortgage rate, the conclusion is the same. If you choose somewhat realistic numbers, the conclusion will still remain the same. For example, suppose we make rent a constant 3% of home value, and make interest rate a low annualized 4%. 401k still comes out well ahead.
Your analysis would only make sense if a person was going to buy a house with cash only. Most people use debt. They take out a 30Y mortgage, so there’s a leveraging effect with your gains. The lower your downpayment, the higher the leveraging effect.
If we assume 20% downpayment and that rent covers monthly expenses exactly for the 30 years (It’ll be higher in reality, but set them equal for the sake of simplicity), we have:
((1.046)^30 / .2) ^ (1/30) - 1 = 10.4%, about the same as the S&P 10%.
There are also 2 things to note:
Rent will most likely make a sizeable profit in addition to the gains from leverage
You can get far more than 4.6% annual gain if you buy in the right areas
Not once you take into account the leverage inherent in home ownership.
Important to note that leverage only exists permanently if the unit is rented. Primary residence houses don’t have leveraged gains after the mortgage is paid off.
For the readers whose eyes are glazing over… worry not. If young folks find a way to save via 401K and/or home ownership, that is a good thing. You can analyze it to death, but in my view the main thing is that they do find ways to save/invest.
I vehemently disagree. Readers should understand how leverage works. Most pseudo-intellectuals I meet tell me that houses are poor investments because they only go up 4% a year while SP500 goes up 10% with no understanding of how leverage makes up the difference.
I also see misconceptions on the other side as well with homebuyers thinking they get leverage with their primary residence when that leverage goes away as you pay down the mortgage.
