More on My Tesla Model 3 Experience After 5 Years

My post describing my experience with the dual-motor long-range Tesla Model 3 I bought back in April 2019 generated further questions about battery life, recalls, and driving range in winter that I answer here.

Battery life

Recently, I was in a waiting room and couldn’t help overhearing a conversation about electric vehicles.  The two people agreed they wouldn’t consider owning one, and the dominant stated reason was fear of high-cost battery replacements.  It became clear after a while that they mistakenly believed EV batteries would have to be replaced as often as lead-acid batteries in gas cars.  The truth is that Tesla batteries are expected to last longer than the rest of the car.

A few days ago, I was startled to receive a notification on my phone that my car’s battery needed to be replaced.  A few seconds of poking around on the Tesla app eased my worries.  I didn’t realize that my car has a lead-acid battery as well as its large high-tech battery.  So, $164 later, my lead-acid battery has been replaced.  The main battery powering my car is still fine.

Recalls

Tesla has been in the news frequently for having “recalls” of their cars.  This brings to mind images of millions of cars lined up at service stations, but that’s not what’s going on.  Almost all of these recall changes are just folded into the next software update.

Tesla pushes out software updates to cars every month or so.  They mostly contain improvements they’ve made.  Sometimes, they include things regulators want.  In one case, the font size on a warning message was made larger.  Quite the recall.

The only substantive recall item I can remember was something to do with cabling near my trunk.  When I was having a camera realigned for US$34, the technician improved the cabling at no extra charge.

Driving range in winter

Electric vehicle batteries have an optimum temperature.  Heating or cooling them draws battery power, and operating the batteries outside their optimum temperature range is less efficient.  There is nothing the driver has to do about this; the car handles it all itself.  However, the net result is that the car’s range is lower in winter than it is during other seasons.  

The difference in range has been noticeable, but not a problem for me.  One time that I looked at some numbers, the difference in range between a summer day and a below-freezing winter day was close to 15%, but I can’t say whether this is typical.  I suspect the difference depends on how cold it gets and the make of electric car.  To be fair, gas cars burn more gas in winter as well.

Conclusion

My conclusion remains the same: I love my Tesla Model 3’s quiet power and its well-designed features that get improved regularly through software updates.  Most of what I see online about EVs is overblown or doesn’t apply to Teslas.  My car is very cheap to run, but the purchase price was high.  It’s been the best car I’ve owned, but until they become cheaper, I can only recommend them to those who can reasonably afford a luxury car.

Retirement Income for Life (Third Edition)

Actuary Frederick Vettese has a third edition of his excellent book, Retirement Income for Life: Getting More Without Saving More.  He explains methods of making your retirement savings produce more income over your entire retirement.  These methods include controlling investment fees, optimizing the timing of starting CPP and OAS pensions, annuities, Vettese’s free Personal Enhanced Retirement Calculator (PERC), and using reverse mortgages as a backstop if savings run out.

This third edition adds new material about how to deal with higher inflation, CPP expansion, new investment products as potential replacements for annuities, and improvements to Vettese’s retirement calculator PERC.  Rather than repeat material from my review of the second edition, I will focus on specific areas that drew my attention.

Inflation

“We can no longer take low inflation for granted.”  “An annuity does nothing to lessen inflation risk, which should be a greater worry than it was before the pandemic.”  “We could have practically ignored inflation risk before COVID hit but certainly not now.”

It’s true that inflation is a potential concern for the future, but it’s wrong to say that it was okay to ignore inflation in the past.  Not considering the possibility of inflation rising was a mistake many people made in the past.  We were lulled by many years of low inflation into being unprepared for its rise starting in 2021, just as many years of safety in bonds left us unprepared for the battering of long-term bonds when interest rates rose sharply.

Inflation risk is always present, and financial planners who have treated it as a fixed constant were making a mistake before inflation rose, just as they would be wrong to do so now.  This underappreciation of inflation risk is what causes people to say that standard long-term bonds (with no inflation protection) are safe to hold to maturity.  In fact, they are risky because of inflation uncertainty.

People’s future spending obligations are mostly linked to real prices that rise with inflation, not fixed nominal amounts.  The uncertainty in future inflation should be respected just as we respect uncertainty in stock market returns.

Maximizing retirement income

Vettese does a good job of explaining that things like CPP, OAS, and annuities provide more income now because they offer your estate little or nothing after you die.  To make full use of this book, you need to understand this fact, and “you have to commit to the idea that your main objectives are to maximize your retirement income and ensure it lasts a lifetime.”

Spending shocks

Retirees should “set aside somewhere between 3 percent and 5 percent of their spendable income each year, specifically to deal with spending shocks.”  “This reserve might not totally cover all the shocks that people … might encounter, but it will definitely soften their impact.”

It’s easy to plug a smooth future spending pattern into a spreadsheet, but real life is much messier than this.  I’ve seen cases of retirees choosing to spend some safe percentage from their savings while also expecting to be able to dip in anytime something big and unplanned for comes up.  This is a formula for running out of retirement savings early.

Retirement income targets

In this third edition, Vettese assumes that retiree spending will rise with inflation until age 70, then rise one percentage point below inflation during one’s 70s, two percentage points below inflation from age 80 to 84, then 1.8% below at 85, 1.6% below at 86, 1.4% below at 87, 1.2% below at 88, 1% below at 89, and rising with inflation again thereafter.

This plan is based on several academic studies of how retirees spend.  I don’t doubt the results from these studies, but I do have a problem with basing my plan exclusively on the average of what other people do.  The average Canadian smokes two cigarettes a day.  Does that mean I should too?

The academic studies mix together results from retirees who spent sensibly with those who overspent early and were forced to cut back.  I don’t want to base my retirement plan partially on the actions of retirees who made poor choices.  Similarly, I prefer to base my smoking behaviour on those Canadians who don’t smoke.

I suspect that Vettese has already tried to take this into account with his assumed retiree spending reductions, because some studies I’ve read show retiree spending declining earlier than Vettese’s plan.  It’s hard to know what would happen to study results if we excluded early overspenders, but I suspect that spending declines would tend to start later and be less severe than Vettese’s plan.

Part of the reason I believe this comes from the fact that societal spending actually rises faster than inflation; it tends to rise with wage inflation rather than regular (price) inflation.  Wage inflation has averaged about 0.75 percentage points more annually than price inflation.  So, even if your retirement plan assumes your spending will rise with price inflation, your spending will decline relative to your younger neighbours.

In fairness to Vettese, though, his free tool PERC embeds other assumptions that are conservative, so the fact that I find this assumption somewhat aggressive doesn’t diminish the usefulness of PERC.

Monte Carlo simulations

“Monte Carlo simulations might be good at modelling investment returns between the 5th percentile and the 95th percentile, but when you go deeper into the tails of the distribution, it is a different matter.”

This is a good point.  When a simulator says your retirement plan gives you a 97% chance that you won’t run out of money before you die, it sounds wonderfully scientific, but it’s just marketing.  Black swan events make it impossible to make such assessments accurately.  Models of investment returns are never completely accurate, and the further we go toward lower probability outcomes, the worse these models are.

CPP survivor pension

The rules for calculating a surviving spouse’s additional CPP payments after the other spouse dies are quite complex.  Vettese explains them well, except that I think one part is inaccurate, as I explained in detail in an earlier post.  

At one point, Vettese calculates a quantity C that is the difference between the deceased spouse’s basic retirement pension and the surviving spouse’s basic retirement pension.  Information I’ve seen elsewhere says that this should really be the difference between the maximum basic retirement pension and the surviving spouse’s basic retirement pension.

In an example for a couple Susan and Nick, after Nick’s death, “Susan’s CPP pension would be increased to the maximum CPP pension payable to any one individual.”  This is because CPP rules don’t permit Susan’s total CPP to exceed this CPP maximum.  However, my understanding is that this applies to the case where Susan’s CPP pension was started at age 65.  If Susan had started CPP at 70, her own pension would still be 42% higher, and when the survivor pension is added , her total CPP pension would exceed what we think of as the maximum CPP pension.  Similarly, if she had started her CPP at 60, her total pension would be less than what we think of as the maximum CPP pension.

OAS

If you’ve lived at least 40 years in Canada between the ages of 18 and 65, you’re entitled to a full OAS pension (that may be clawed back if your income is high).  However, if you haven’t been in Canada for that long as an adult (but for at least 10 years), your OAS pension “will be reduced in proportion to the number of years of residency … between ages 18 and 65.”

If you delay taking OAS until somewhere between age 65 and 70, the extra years can be used to either increase your years of residency closer to 40, “or to adjust [your] pension upward by 7.2 percent for each year beyond age 65,” but not both.

Deferring CPP

The best reasons for not deferring CPP to age 70 are that either you don’t have enough savings to bridge the retirement gap before age 70, or that your health is severely compromised making an early death certain.  Beyond that, Vettese explains why most other reasons people give are irrational or invalid.

“Deferring CPP pension to age 70 forces you to draw down your [assets] more quickly before age 70, but those same assets last longer because the CPP pension from age 70 and on is so much bigger.”

This is an important point that I find people have a hard time understanding.  They imagine having to spend down their assets, and then imagine a bleak future.  In reality, they will have higher guaranteed CPP income that will insulate them somewhat from stock market gyrations, and their assets will ultimately last longer.

Most of the book’s analyses are based on getting below median investment returns.  In one chapter where Vettese looks at what happens if investment returns are at the median, he concludes that deferring CPP to age 70 “makes eminent sense whether future returns are poor or middle-of-the-road.”

Reverse mortgages

“With a reverse mortgage, you do not have to make any payments while you or your spouse continue to live in the home.  And you cannot be forced to move out.  You do have to maintain the home, however, including paying property taxes and home insurance premiums on a regular basis.”

This is self-contradictory.  You can be forced to move out if you don’t maintain your home or fail to pay property taxes and insurance premiums.  We tend to think of home maintenance as an annoyance, but something we can handle.  However, I’ve known several old people who simply couldn’t maintain a home properly.  Many old people rely on neighbours and family for help with basic maintenance, and many others simply fail to maintain their homes properly.

We are still in the early stages of Canadians taking on reverse mortgages.  Financial institutions are unlikely to deliberately generate bad press while they are trying to make sales.  But in a decade or two, when they have many clients who owe as much as their homes are worth after the cost of selling, the mood will change.  Financial institutions will be incented to force old people out of their homes if the homes are run down.

PERC built-in assumptions

Vettese explains the assumptions built into his free PERC tool, including that “the primary spouse (the one entering the data) is assumed to die at 90 and the surviving spouse at 93.”  However, in an earlier example, “Graham dies at age 85, the default assumption that is built into PERC.”  I’m not sure if this is an error, or if these are somehow built-in assumptions for different types of cases.

For anyone who finds assuming death at 90 or early 90s not to be conservative enough, PERC also assumes that you want to be left with 10% of your assets remaining at this age, which is on the conservative side.  So, depending on what different assumptions the reader might make, these differences might balance out somewhat.

Another thing to keep in mind is that PERC displays income in pre-tax amounts.  I’ve always done my own planning with after-tax amounts, so it takes some work to compare my outputs with those of PERC.

Decumulation strategy

“Retirees are generally better off drawing down assets from multiple sources on an annual basis (in proportion to how much they have of each asset) rather than drawing down taxable assets first and trying to keep their tax-sheltered assets intact for as long as possible.”

This agrees with my own simulations.  However, I’ve found that it’s generally better to draw down taxable assets before drawing down a TFSA, particularly if the taxable assets don’t have large unrealized capital gains.  It’s RRSPs and RRIFs that are best drawn down gradually throughout retirement.

Financial advisors

Controlling investment costs, deferring CPP and OAS, buying annuities, and even planning to rely on a reverse mortgage all cut into financial advisor compensation.  “I know a growing number of financial planners out there who do accept the ideas in this book,” but Vettese also has stories of advisors whose recommendations help themselves rather than their clients.

Conclusion

This is an excellent book about the complex task of decumulating assets in retirement.  The tone of this review may seem negative, but this is because I included every case where I disagreed with the author.  I highly recommend this book to anyone who is either retired or close enough that they need to consider how to create an income from their savings.

Tesla Model 3 Experience After 5 Years

Back in April 2019, I bought a dual-motor long-range Tesla Model 3.  I had some concerns about buying an electric car, but my research showed that none of these concerns were likely to be real problems.  Still, buying the car felt like a leap of faith.  Now that I’ve had it for 5 years and 4 months, I can say that it’s been the best car I’ve owned, even better than my Lexus GS400.

Some concerns I had before I bought the car were its price, the charging network when traveling, home charging, battery reliability, self-driving features, and ongoing costs.  Something I didn’t know about in advance that turned out to be a pleasant surprise is the one-pedal driving.  A negative about owning this car is that somehow EVs have become political.

Price

Compared to other luxury cars, Teslas are reasonably priced.  However, there is no lower end Tesla model suitable for the bulk of car owners.  Teslas are still too expensive for most people.  The lower annual cost of running a Tesla compared to gas cars helps to mitigate this problem somewhat, but isn’t enough to solve it.

Charging network

There is a huge difference between the Tesla and non-Tesla charging networks.  Tesla completed full coverage of the continental U.S. and Canada with chargers a few years before I bought my car.  I’ve never had trouble finding a charger when I needed one.  They’re generally very close to major roads and highways.  About 95% of the time, when I arrive at a charging station, there is a working charger available for me.  The other 5% of the time when the chargers have been busy, I’ve only had to wait 5 or 10 minutes to start charging.

The non-Tesla charging network is a different story.  While much of what you see on social media about EVs is nonsense, the problems with the non-Tesla charging network is real.  Quebec is a notable exception where they’ve gone to a lot of trouble to build a decent network.

When traveling, it’s common to charge from about 20% of battery capacity to about 80%.  Just as with a gas car, you generally don’t run your car to empty.  However, stopping a “fillup” at 80% of battery capacity is different from how we run gas cars.  The reason is that the charging rate slows down a lot as you get nearer full charge.  So, it’s faster to charge up a little more often than it is to charge to the top each time.

On average on long trips, I charge for about 20 minutes every 3 hours.  It was a little slower 5 years ago, but newer Tesla superchargers are faster.  Superchargers are nowhere near as plentiful as gas stations, but my car tells me when to stop and it directs me to the chargers.  

My wife and I are getting older, and we need the 20 minute stops to stretch our legs and find a bathroom.  Extending what would have been a 10-hour trip with a gas car to 11 hours with our EV has been good for our mental and physical well-being.  Others may feel differently.

Home charging

A fact that some people have a hard time wrapping their minds around is that people whose homes have garages hardly ever have to use the charging network.  I had a 220V, 48A charger installed in my garage.  I just leave the car plugged in whenever I’m home, and the car’s fancy software figures out the best way to charge up and stop charging when no more range is needed.

My charger can add about 60 km of range for each hour it’s plugged in, but I rarely notice.  Just as you don’t stand there watching your washing machine spin, there’s no need to watch a car charge.  Each day I wake up to a topped up car.

It’s possible to charge my car with a regular 110V outlet.  At 12A, an hour of charging adds about 7 km of range.  This isn’t much, but plugged in for 10 hours overnight when visiting friends out of town adds about 70 km.  When away from home, I’m often able to stay sufficiently charged for day trips using just a regular 110V outlet.  On a 5-month stay at a rental home in Florida, I only needed superchargers 3 times other than for the long drives between Canada and Florida.

Battery reliability

There’s no question that it would be expensive to replace the batteries in my car, just as replacing a gas car’s engine is expensive.  Fortunately, batteries don’t tend to suddenly fail.  They degrade.  My experience seems to be fairly typical: my range has declined from 500 km to 460 km, or about 8% in a little over 5 years.

This loss of range was faster initially, and is about one percentage point per year now.  If this persists for another 10 years, I can look forward to a 15-year old car that still has a range of 410 km, which is more than usable.  It might work out better or worse than this projection, but the odds are that this car will be scrapped for some other reason before the battery gives out.

Self-driving features

My car can drive itself in most situations with the latest software upgrades, but the truth is that I generally only use these features on highways for now.  Having the car drive itself for long highway stretches greatly reduces mental strain, particularly in the dark.  A day of driving used to leave me wrecked, but not any more.

For short distances I don’t feel the need to use the car’s self-driving features.  When I control the car myself, the safety features are still enabled.  A couple of times now when I’ve needed to make a sudden stop, and the car began braking before I could get my foot to the brake.

The most important features to me are the safety features, keeping the car in its lane, and adaptive cruise control.  Many other EVs and gas cars have these features as well.  Talking to owners of other cars, Tesla’s technology appears to be better, but perhaps there are cars I have no information about that have good self-driving features.

Ongoing Costs

My Tesla has never had or needed regular service.  All my gas cars have needed oil changes and other fluid top-ups a couple of times per year.  Even my brakes last longer because my car usually slows down using regenerative braking.  My brakes haven’t needed any repairs so far.

Apart from replacing tires once and swapping summer and winter tires, the only maintenance I’ve done is new cabin air filters for C$67, a camera realignment for US$34, and windshield washer fluid.  I don’t get charged anything extra for the software updates my car gets every month or so.

However, the big savings come from the low cost of charging vs. the cost of gas.  My car has averaged 0.185 kWh per km driven.  At the 13 cents per kWh I pay on my hydro bill, this works out to only 2.4 cents/km.  Tesla charges double to triple this amount at their superchargers, but I don’t use them enough to make much difference to my overall costs, and supercharging costs still end up being cheaper than gas.  Some hotels offer free EV charging which saves some money on long trips.

If my Model 3 were a gas car, Tesla says that it would get about 30 MPG, or about 7.8L/100km.  I’m guessing that a comparable 350 horsepower gas car would use premium gas.  The current Costco premium gas price in my area is 174.9 cents/L, and my car as a gas car would cost 13.6 cents/km to run.  So, I save 13.6 ‒ 2.4 = 11.2 cents/km.  For the 85,000 km I’ve driven so far, I’ve saved about $9500.

One-pedal driving

I didn’t know about one-pedal driving before I bought my Model 3.  When you ease up on the accelerator, it doesn’t just coast the way a gas car does; the car does some regenerative braking.  So, for most driving, including cornering, you just need to use the accelerator to speed up and slow down.  I only need the brake pedal for panic stops and coming to a final stop after slowing down to about 10 km/h.

It took a few days to get used to one-pedal driving, but now I really like it.  Whenever I drive someone else’s car, I find the going back and forth between pedals annoying.

Politics

Most people seem to find Teslas interesting, and they have questions.  However, a minority react negatively to my car for political or ideological reasons.  This isn’t a problem when it’s just words, but I occasionally get other vehicles doing stupid and dangerous things around me.  Fortunately, this has declined a lot since the first year or so I had the car.

Other people react negatively because they see it as a show off car, like a BMW.  It may be that some others buy Teslas for these reasons, but I don’t even wash mine.  I’ve never seen the point of washing cars.  I tend to think more about practical matters.

Conclusion

I love my Tesla Model 3’s quiet power and its well-designed features that get improved regularly through software updates.  It’s very cheap to run, but the purchase price was high.  Overall, it’s been the best car I’ve owned, but until they become cheaper, I can only recommend them to those who can reasonably afford a luxury car.