One of the major decisions when investing is whether to invest in retirement accounts or in a standard taxable brokerage account. There are so many factors playing into this that it is impossible for me to consider every imaginable situation so I will try to lay out some of the considerations and then I hope you will add your own thoughts or questions in the comments 🙂 For those of you that just want a quick answer I made this decision cheat sheet:
I also created one for “ratepension” that is a bit more complex and still doesn’t give clear answers – that will be the focus in today’s article:
Regarding “livrente” I might have oversimplified it a bit but generally it’s more like an insurance than a retirement account and the payouts are probably too low compared to your investments unless you live far longer than average – the retirement companies won’t even give data on it. I will expand on it in a future short article but the conclusion in the interim is to never willingly put extra into livrente (I can’t stop my employer contributions and that’s fine, but I’d never add to them).
On the other hand, in an “Aldersopsparing” you have full control and access to cheap ETF’s – and if you have other stock investments you’ll usually also save on taxes since it will be easier to stay below the stock tax progression limit. I’ll also do a future in depth article on Aldersopsparing soon but I already shortly mentioned some benefits in a previous article. The trickiest decision though is whether to fund a ratepension if your employer doesn’t already max it so let’s dig into that.
Basics of ratepension
Let’s start with the basics to get on the same page. This retirement type is funded with pretax income – except you actually still pay 8% job market contribution from your salary before funding the account. You can only contribute up to 53,500,- (so 58,152,- before the 8% tax) a year – this used to be 100,000,- a few years ago and unlimited before that. At the other end of the deal the money in the retirement account will be paid out in monthly rates over 10-25 years as instructed by you. You can start taking payments up to 5 years before “folkepensionsalderen” (age of eligibility for public pensions) and you must have the last rate paid out within 20 years of folkepensionsalderen (so that’s where we get the 25-year max). Alternatively, you can in principle roll it over into livrente if you so please, but unless you are in quite exceptional health and have a family history of longevity you probably shouldn’t.
Tax implications of ratepension
The idea that is oft quoted when ratepension comes up is that if you are paying top level income tax now you can pay into ratepension with pretax income and avoid that last ~15% of taxes now. Then you can pay out the money from the retirement account in your retirement years while paying ~15% less taxes. This benefit has been quoted by banks, retirement- and insurance corporations and has pushed a lot of people in my parents’ generation to fund their ratepension with 100,000,- or more each year but unfortunately in some cases they end up with less.
The culprit with regards to ratepensions-payouts is the structure of the public pension in a base part and a part conditional on income. Actually, all of the pension is conditional on your income but the base part is only tied to employment income (anything that you pay 8% job market contribution on) and only starts to kick in if your employment income is above 316,200 kr (2017) after the 8% tax. So, it is a non-issue to most seeking FIRE and even if it affects you then that’s probably a sign that you will have more than enough. The part that affects a lot of us though is that “pensionstillægget” (the part conditional on any income) is lowered already when you earn 69,800,- (2017) or more in each calendar year from retirement account payouts or other non-employment income (140,000 for couples). When you earn above this limit then every single kr you earn pretax reduces your pretax pensionstillæg with 0.309 kr. until it is completely gone at 324,000,-. So, let’s look at an example to get a grasp on the potential price you will be paying for filling up your pretax retirement accounts.
Cheap Camilla and Proactive Peter
Cheap Camilla knows how to live frugally and doesn’t need all that much in retirement. Instead of working far too long to fill up her ratepension she quits her job early and has just enough in there to stretch it as 69,800,- a year over her 25 years of eligibility. So during retirement she receives:
- Yearly payout from ratepension (+livrente+atp etc.): 69,800 kr
- Yearly payout from folkepension: 73,900 kr
- Yearly payout from pensionstillæg: 78,600 kr
This works out to a nice 222,300 kr. pretax and with an average municipality tax and no church tax we have 156,000 net (see my tax calc). This is 13,000 net a month which is more than a lot of unemployed people live on and quite manageable for a lot of us in paid off homes. Do note though that pensionstillæg is lowered for couples so for a couple the equivalent amount would be 263,000,- net or 22,000,- a month, so not quite double.
Proactive Peter worked his whole life to gather money, same as his parents before him, and because he paid top level income tax he stashed away quite a large ratepension to save on taxes. Even when stretching it over 25 years he still receives a payout of 169,800 kr a year – a clean 100,000 more than Camilla. So, what does he net:
- Yearly payout from ratepension (+livrente+atp etc.): 169,800 kr
- Yearly payout from folkepension: 73,900 kr
- Yearly payout from pensionstillæg: 78,600 – 100,000*0.309=47,700 kr
This works out to 291,400 kr. pretax and with an average municipality tax and no church tax Peter can take home 199,000 kr. So, while having an extra 100,000 of income per year he only had an extra 43,000 kr a year paid out or an effective marginal tax of 57%. How crazy is that?! Also, keep in mind that the money in his ratepension was already taxed at 8% on contribution so he is far above the official tax-ceiling (tax ceiling is 52%+church of about 0.6% after 8% amb).
This illustrates the main issue with funding your ratepension. At best, you fund it with money that would have been taxed at around 52% and you end up taxed at 57%, at worst you added about 20 percentage-points to your marginal tax. Still we’re just getting started on the complexity and Peter might still come out ahead since:
- He can structure his income in a way to reduce total cuts to pensionstillæg
- He has had some growth at 15% PAL-tax rather than 27-42% (only an advantage in shorter timeframes)
- He might have had access to cheaper funds (especially ETF’s) inside of the retirement account
Taxes on retirement ratepension
We have 3 breakpoints for taxes on the payouts from your retirement accounts. For someone not earning any income aside from those payouts and folkepension they will look like:
- 0-69,800 kr: Normal bottom bracket income tax
- 69,800-324,000 kr: Normal income tax + cuts to pensionstillæg
- 324,000-406,000 kr: Normal income tax
- 406,000 kr and up: Top bracket income tax
The 406,000 comes from the progression limit at 480,000 kr minus the folkepension at approx. 74,000 kr.
Consequently, the marginal and effective tax we pay in an average municipality would be:
- 0-69,800 kr: Marginal 37.4%, Effective 37.4%
- 69,800-324,000 kr: Marginal 56.7%, Effective 37.4-52.5%
- 324,000-406,000 kr: Marginal 37.4%, Effective 52.5-49.5%
- 406,000 kr and up: Marginal 49.5%, Effective 49.5-52%
Also, note that if we only look at the part of payouts above the 69,800 kr then the effective tax on that will always be above 52% and often above 53% so higher than top bracket income tax.
Growth in retirement account
First just a simple check. Say we pay in the same amount (corrected for taxes) to either a ratepension or a normal taxable brokerage account and say that the fees & the growth is similar, and that there are no dividends. Let’s also for now say that the marginal tax on the contributions to ratepension equals the effective tax on the payouts and finally let’s assume a zero inflation. Those are big and generally bad assumptions but they are only meant to illustrate a point – growth at PAL-tax is only superior to 27% tax on timespans shorter than 25 years if we assume 7% growth:
Of course, this only holds if you can actually keep stock taxed at 27% rather than 42%. I’ll look at stocks and taxes another time but you might stay close to the 27% progression limit even on even large amount of stocks since you are only taxed at realized profit and since dividends will keep your cost basis up even when they are otherwise annoying. Also, the graph is similar at 35% effective tax on stocks except the difference is insignificant for 50 years instead of 40.
One way to think of this is that if you are looking at one year returns then obviously a 15% tax is better than a 27% one. But the 15% tax hinders compound interest so it’s a race where the Pal-tax starts out ahead but over long time frames nothing can keep up with compound interest. I was surprised when I saw that the catchup time frame was only 25 years under the given assumptions and it certainly limits the advantage of utilizing retirement accounts during youth.
Here is the sheet with the two above graphs – make a copy and change assumptions or check calculations if you want, but there is a more refined sheet just below: Graph Sheet
So, can it be worth it
I made a sheet that calculates outcomes on investing your pretax income into a ratepension versus just a standard brokerage with a few more input parameters. You can find it here: Calculator: Ratepension or Taxable Brokerage
It still doesn’t consider the specific payout schedules you can take advantage off to limit cuts to your “pensionstillæg”. Playing around with it for a bit though, you’ll see that if you are not taxed in the top income bracket currently it is almost never worth it to pay into ratepension. Also, you might note that if you’ll end up paying around the same marginal tax on payouts from the retirement account then it’s only worth it for relatively short timeframes and even then, it doesn’t make a big difference.
Alternative payout schedules & delayed folkepension
Without going too much into depth (that would be yet another big complexity best reserved for another time) there are a lot of considerations regarding when to take payouts from your ratepension. There are four basic approaches you can take:
- Pay out over 25 years to lower the yearly amount and thus minimize taxes each year.
- Pay out over 10 years starting 5 years before folkepensions-age. This way you will probably cut your entire pensionstillæg but only for 5 years instead of 20.
- Try to hit a sweet spot between the two above. The marginal tax between 324,000-406,000 kr is only 37.4% so rather take 406,000 kr for a few years less than hit 324,000 kr every year.
- Pay it all out at the 60% penalty
Generally, I think you’d take the first approach if your account holds around 1.3 million kr or a bit more. If you have substantially more than that then option two and three would probably start being more attractive. Option four is not attractive in realistic circumstances (also not possible in many accounts) but is an emergency brake in case of new laws that taxes your retirement funds even further – I don’t expect politicians will want to do that though.
Another more unconventional option is that if you are still working you can delay your folkepension by up to 10 years to earn up to 35% extra on the payouts. This is generally not worth it for average expected lifespans but if you do work the required 1000 hours a year and your pension payouts would be heavily taxed if you took them then you might want to consider delaying them.
Additional spending but still full pensions
Initially in this blog-article I noted how 156,000 of net income was fairly optimal for tax reasons since additional income is taxed heavily through cuts to your pensions. However, you can spend more than that if you want in a few ways. The obvious is that any savings you have won’t count as income so you can spend from those. Aldersopsparing is an example of this and is a fine way to keep your savings growing without filling your 27% stock-tax progression limit. But you can also sell stocks for higher amounts than the 69,800 pension limit as long as the realized gains lie within that limit. For new investments or investments that otherwise have a high cost basis (from dividends or tax harvest) this can mean big withdrawals without much tax-footprint that deducts from your pensions.
Other than that, there are two smaller deductions aimed at retirees:
- You can earn 10,000 kr tax free if working in private homes
- The first 60,000 kr (after 8%) employment income you earn isn’t counted anywhere when calculating your pension payouts.
government subsidies per spending
Till now I’ve talked about how cuts to your pension payouts is effectively a tax on your income. One could also argue that the public pension is not yours in the first place and that you should just count yourself lucky to get it. That discussion can easily get political so I will abstain here but I feel like that point of view merits at least a mention. A graph could be made that would basically be an inversion of the effective tax graph above but I’ll have to owe you that 🙂
That was a long post and it was mostly for nothing. The actual advice does not differ much from what you could probably have found in a lot of other places. finanshus.dk says it like this:
- Invest in an “aldersopsparing” type account if you don’t need the money right now
- Invest in other retirement accounts only if you pay top bracket income tax now and will not pay top bracket income tax in retirement
I do like to do the calculations myself though rather than take advice online at face value and I hope this post has helped shed some light on the subject for you – it did for me. I also hope some of you will post your own comments, plans and questions here.
I know it’s been a long hiatus without any posts but it’s hard to find the time to research, create spreadsheets and write these long posts. I will try to post more frequently but I’ve also been considering an outlet for shorter articles or links to interesting new articles etc. I know a Facebook group would be the usual solution for that, but I don’t really like Facebook. I’ve been considering a forum, a chat, a wiki or a subreddit – check the comments here: buy-or-rent and offer up your input either there or on this post.