The Ermine is advancing of years. It comes to us all, hopefully. I am not yet of the age when I would have quit The Firm after thirty years of service and gone to the pub to celebrate my forthcoming freedom. But it’s not that far off.
Seven years after retiring, I have now burned through half my DC pension AVC savings and invested the other half into the ISA. In cash-flow terms the Ermine is almost skint, and I am on the final approach to taking my main pension, a little early. So I asked them for the information pack.
Every pension is different, and with mine there is the usual option to take some as a tax-free lump-sum. After that there are two options. One is to take a pension that is index-linked up to a cap, and the other option is to take a higher pension that is part non-increasing, and part index-linked up to the same cap.
Why on earth would you do that? Well, the logic is that you get a higher start, since they are paying you to take your claim of index linking off their hands. The higher start makes some sense – one is in best health at the start of retirement than at the end when you’re eyeing up a pine box. They also made something of the fact that the State Pension will turn up later, which should fight some of the later attrition.
Inflation can easily kill the higher start
It looked reasonable at first sight. There’s a common argument that people spend more in the early years of retirement, and less as they get older. Unless they are unlucky enough to end up in a care home, but fewer than 20% of us end up in a care home1. The difference between the higher start and regular isn’t huge, however, it is about three grand p.a. after tax. But it doesn’t take much inflation to erode that, it falls behind in my early seventies and typical inflation rates of ~3%. Some people take the argument that it is the cumulative shortfall that matters,and it is true that this falls below zero later on, a few years past the age that my Dad died. But I am in much better health than either of my parents were at my age. So I might be leery of assuming I cark it at the same age, though of course I could be flattened by a bus tomorrow. Risk is tough to qualify, eh?
What is also bad is this erosion is progressive, and erodes Mrs Ermine’s pension benefit from this after I am gone. The higher start is the theft that keeps on taking, and since there is a significant age difference this loss compounds for longer. I had to run the simulation to 100, not particularly because I believe I will live to 100, but it’s reasonable to anticipate that Mrs Ermine may still be living on what would be my 100th birthday
Another option is to take a lump sum. This reduces the pension, but I am not looking to max this, the commutation rate is 26% post-tax2 which is good though not fantastic. However, I could invest this into my ISA for a few years, with a quicker start because I can repay into Charles Stanley what I borrowed from my ISA, because without an income you are a pariah to the financial system. So I had to borrow from myself. The lump sum can pay back that withdrawal and take out this year’s ISA allowance, and I can then fund some more years.
There are other ways to smooth income
I concluded that the principle of the higher start is good, but giving up my inflation-linking isn’t. There are other ways to do this. I will run down some of the lump sum at about 3k p.a. to augment my income. I will invest into my ISA to the maximum allowed, from the lump sum. When the lump sum falls to zero I will then start to take an income uplift from the ISA of about 3k p.a. until I get my State Pension. The numbers aren’t set in stone – I could take about 6k which is the difference my SP would make (I pay tax on all of it due to having existing income) and/or vary it with what I am doing that year.
Beware the inflation monster
I was surprised at the difference inflation made. Particularly inflation early on. What could cause high inflation in the near future? Well, even Brexit fans seem to think it will bugger the economy. When I say high, I only modelled inflation of 5% for a couple of years, followed by inflation of 2% p.a. for the rest of my life. We aren’t talking the near~27% inflation that has been known in my living memory. In an ideal world I would Monte Carlo simulate this. I know how to do that for circuit design in filters, but I am not clever enough to make Excel do that3. It’s probably possible using something like R, but I would be drawing the pension by the time I learned how to wrangle that. Fiddling with the values in Excel shows me that most of the danger is early on, in the first five-to ten years.
Inflation is tough to model in Excel. You can, of course, apply the compound interest function FV. Good luck with that if you are trying to simulate varying interest rates over the years. I chose to use the gonzo method, which is to track what inflation does to £100. So in any one year, if inflation is 3% then you need 100 * (1+3%)=£103 to buy the same amount of stuff the next year. My years are in columns, and I repeat this exercise each column to show what I need to have £100 in value after cumulative inflation.
I then divide the non-inflation adjusted amounts by the inflation adjusted value fo £100 and multiply by 100.
I confess that seeing this makes me wonder if all those people taking CETVs of DB pensions have balls of steel. Big numbers can turn into smaller numbers counter-intuitively quickly. Or maybe I am just a pussy. This is a fight I don’t want to fight as I get older, even though I have experience of investing successfully.
The ISA has grown well over the last twelve years and I thought I was going to burn out my SIPP savings in three years on living costs. As it was I used a fair amount in a house purchase and paid more tax on it as a result of that purchase, but investment gains meant that these extra costs and some ISA contributions were covered. In some ways the experience of investing makes me more risk-averse with the pension, after all markets are high and there is plenty of hazard. I have an ISA for that racy stuff, and the principles of diversification point to choosing a balancing asset class with different characteristics. An annuity, which is what a DB pension is, has a very different risk/return profile to an equity portfolio.
With a reasonable stroke of luck if I get to the age my parents got to I could end up retired for longer than my working life. Even modest amounts of inflation look like a bastard over that sort of timescale.
If you’re fortunate enough to be offered a higher DB pension start in return for surrendering inflation protection, then look that gift horse very carefully in the mouth. At least consider the alternative of taking a bigger lump sum and subbing yourself until you get the State Pension. I do not need to take more risk to get an adequate income. So I won’t. The words of Warren Buffett on the clever sods at Long-Term Capital Management still echo across the years, though they were uttered when I was a young pup trying to make money in the dotcom boom4.
…But to make money they didn’t have and didn’t need, they risked what they did have and did need, and that’s foolish. That is just plain foolish. Doesn’t make any difference what your IQ is. If you risk something that is important to you for something that is unimportant to you, it just does not make any sense. I don’t care whether the odds are 100 to one that you succeed or 1,000 to one that you succeed. […] And yet people do it financially without thinking about it very much.
There’s still truth in those words. I have no human capital left, so I need to be more careful with my financial capital. The guys from LTCM had the advantage of being young enough to be able to start again5.
- https://www.mha.org.uk/news/policy-influencing/facts-stats/ “Approximately 416,000 people live in care homes. This is 4% of the population aged 65 years and over, rising to 16% of those aged 85 or more.” ↩
- The lump-sum is tax-free so the commutation rate must be calculated on the net of tax value. I don’t give a shit about money the taxman gets, it’s no use to me. This lifts the commutation rate a few percent as tax-free lump sum is more valuable than taxed income. Up to a point, because more tax-free cash also increases my risk profile as some needs to be invested to get an income. ↩
- A few years ago I did try a Monte Carlo simulation using Excel in this post. You have to run multiple parallel traces of data and use the RAND() function. It was enough to convince me that Excel was the wrong tool for the job, and there is limited use in an Excel spreadsheet that gives you a different answer each time you open it due to RAND(). ↩
- I did make money in the dotcom boom, despite churning, having a cavalier disregard for dividends or indeed fundamentals. It was losing it in the bust and quitting around the low-water mark where I went wrong. The training for my future self was cheap at the price ;) ↩
- though it appears that some of ’em didn’t learn from the experience since they went titsup again. Shoulda listened to Warren…. ↩