Three weeks before Russia invaded Ukraine, the highest-ranking military officer in the United States, General Mark Milley, predicted that Ukraine would “fall within 72 hours”.
This was the conventional wisdom ahead of the invasion, yet something everyone knew to be true was proven to be very, very wrong.
Milley testified before congress again in early April updating his prediction, “I do think this is a very protracted conflict, and I think it’s at least measured in years. I don’t know about a decade, but at least years for sure.”
How could America’s most senior military advisor have been so wrong?
US intelligence agencies are still searching for answers, but a key factor was certainly the limited amount of data on Russian military conflicts. Russia hasn’t fought a major land war since the 1980’s.
How Much Money Can I Spend in Retirement?
Conventional wisdom typically starts with the 4% rule. Specifically, this is the idea that you can spend 4% of your initial portfolio value, adjusting that spending every year to account for inflation and have a safe and secure retirement.
Like General Milley’s Ukraine prediction, this conventional wisdom comes from a limited amount of data on one country in particular.
A 4% withdrawal rate has worked very well in the United States. An allocation of 60% stocks and 40% government bonds has only run out of money after a 30 year retirement a handful of times. Failure rates for 30 year retirements, starting every month since 1871, are shown here:
However, US Investors have had a remarkable run. Since 1871, the US had higher stock and bond returns along with lower inflation than almost all other countries.
Failure is not an Option
Adding 15 other countries to the graph above provides us with the dramatically different experience of investors outside the US.
Ouch.
Unlike in the US, retirement failures were common in other countries when using a 4% withdrawal rate. A 60/40 stock/bond allocation ran out of money in a third of the starting periods.
A 4% withdrawal rate clearly seems riskier than the conventional wisdom would have us believe.
You don’t want your financial advisor at the podium a few years into a downturn coming to you with this:
Does Lowering the Withdrawal Rate Help?
Here are failure rates using a 3% withdrawal rate for a few, mostly equity portfolio allocations:
Lowering the withdrawal rate to 3% helped substantially lower failure rates. The UK, the Netherlands, Switzerland, Australia, Denmark, and the US all have failure rates under 1% for balanced portfolios.
But failure rates were still surprisingly high.
We also see a clear pattern that higher stock allocations lead to lower failure rates. On average, a 100% stock allocation had a 16% chance of failure while a 60/40 allocation had a 20% chance of failure.
Let’s lower the withdrawal rate to 2%.
Most countries have no failures, but even with a 2% withdrawal rate a handful of countries still had uncomfortably high rates of failure.
Feeding the Beast
The Netflix documentary series Tiger King follows small time zoo owner Joe Exotic who has an obsession with tigers and fame. Joe, who is eccentric to say the least, discovers that cute tiger cubs help sell tickets to his zoo and attract attention to himself. As the tiger cubs grow into adults, they also grow into their adult appetites. Desperate to feed them, Joe resorts to feeding them horses and expired meat. The series seems to imply that Joe eventually resorts to killing some of the adult tigers to spare himself the expense.
Like Joe, retirees have to feed the inflation beast. If the inflation beast gets too much of an appetite, it eats your retirement dreams alive.
This is precisely what happened to retirees in the handful of countries with high failure rates. Even using a 2% withdrawal rate, retirees had their dreams eaten by the beast.
The chart below shows how many of the historical 30 year starting periods had annualized inflation of more than 10%.
There is a clear relationship between retirement success and inflation. After 30 years of 10% annual inflation, the beast has dwindled your purchasing power down to only 4% of its original value. Before even considering withdrawals!
Withdrawal Rates for Globally Minded Investors
Investors now have the option to diversify globally. Failure rates for global investments into stock and bond markets, equally weighted across all 16 countries, are much improved.
To get to the near certainty people have come to expect from withdrawal rates given the US data, a 2% withdrawal rate for a mostly equity portfolio seems to be a better benchmark for an investor who is globally diversified.
Of course, we all have to decide our own level of risk tolerance, along with our need and ability to stomach that risk. Requiring certainty causes an investor to save far more for retirement than will likely be needed.
Perhaps investors should be more comfortable with some level of failure. This time period involved multiple financial panics, two world wars, a great depression, and periods of sustained high inflation. It might make sense to question how likely these worst case scenarios are to repeat. With the advent of nuclear weapons, if we have another world war, my concern is unlikely to be my investment portfolio.
On the other hand, a globally diversified portfolio might paint a more rosy picture than is realistic as equity and bond market correlations have increased with the advent of globalization. Due to this, global diversification might not be as helpful as it was in the past.
Personal finance is personal and we all have our own risk preferences. This research should highlight that there are no guarantees in financial markets. You can use the Withdrawals Graph to make decisions specific to your situation and asset allocation.
The Risk of Bonds to Your Retirement
It seems counter intuitive, but 100% stocks seems clearly optimal.
Using our 16 country sample of three different withdrawal rates, we see that a 100/0 stock/bond allocation is optimal (has a lower failure rate) 79% of the time, compared to a 60/40 allocation.
Moreover, choosing a 60/40 allocation has resulted in failure rates that are, on average, 8% higher than a 100/0 stock/bond allocation.
Even in the cases where a 100/0 stock/bond allocation was not optimal, it wasn’t much riskier. In the 21% of instances where a 100/0 stock/bond allocation was not optimal, the failure rate was only 2% higher on average.
However, a 100% stock allocation might be harder to stick to psychologically. It will be much more volatile in the short term. Even if it is more mathematically optimal, if you can’t stick to the strategy it will not be optimal for you.
Thoughts?
I’d like to hear your thoughts. Let me know what you think at chris@factorgraphs.com!