This is a somewhat hypothetical question, but something I need to plan for now-ish.

The general advice in Sweden is to keep 3 months of mandatory expenses in a cash buffer with instant payout and zero fees, such as a savings account. I guess this advice holds for any country with universal healthcare and good unemployment protection.

Any larger "rainy day" fund risks getting expensive due to inflation and sub-inflation interest rates on savings account. Hence, it is advised to keep any extra savings in funds or stocks.


So let's assume the day has come to touch those extra savings. Let's assume that there is no urgency but the topics "new car" or "buy a house" are on the table. Time to slowly convert those funds and stocks into cash.


Is there an "algorithm" to determine which funds and stocks to sell?

Please note that I don't mean a firesell here. I have maybe as much as 3 months to carefully plan and time exits.

I am looking for something like "sell least volatile first". Or "sell high-return since purchase first, then least volatile".

Non Answers

I googled around and there seems to be a plethora of advice when to exit an investment. However, most of these strategies aim to maximize ROI of an infinite investment portfolio, and don't specifically address the issue of exiting due to needing cash.

  • 2
    Just sell a percentage of each investment so your overall allocation stays the same. It doesn't need to be complicated
    – minou
    Commented Apr 6, 2021 at 11:30

5 Answers 5


Unfortunately there is no perfect answer for this. First, there is a huge difference between liquidating 5000-10000€ for a "new" used car, 20-30k for a new car and >>100k for a house/apartment. Buying a new car can be pretty urgent if you have a crash, so you may not be able to get a 3 months warning. OTOH if you just need something to drive to work or your hobbies, a cheap used car can be financed from your emergency fund. Buying real estate is likely to liquidate almost all your investment and therefore the most troublesome, but it has also the potentially largest time in advance.

Second, any form of liquidation effectively times the market. You might sell your bonds (because they are less volatile) and see the stock market drop next month. You might sell stocks (because they are at an all time high) and see the stock market continue to soar while you are left with your low-performing bonds. Therefore any liquidation should keep your asset class allocation intact to ensure you are still matching your risk-return preference.
Everything else is fine-tuning and subject to your specific portfolio and taxation laws. For example, it can make sense to realize both gains and losses to minimize the taxable gains.

Third, there is the timing issue. Just like entering the market where you are faced with a lump sum investment versus cost averaging decision, you are now faced with the decision between a one time sell off or selling in chunks. Depending on how markets develop, you might be better off with one or the other decision. As markets tend to go up over time, there might be a slight edge for a late one time sale, but most people are probably more comfortable with selling multiple chunks over a longer period of time.

  • Awesome answer! It both highlights the complexity of the decision and gives a rule of thumb. ("liquidation should keep your asset class allocation intact") Commented Apr 6, 2021 at 11:54

You make an assumption that the non-rainy-funds are relatively illiquid.

What's common is to have a targeted mix of assets, often simplified to Stocks/Bonds (Yes, one can get very fancy, Stocks/Bonds/Metals/RealEstate/Crypto. For this example, let's keep my response simple).

Say my mix is Stocks 70% / Bonds (note, bonds include CDs, money market funds etc) 30%. When I need cash, it naturally comes from the Bond/Cash assets. This at least delays the need to pick what to sell from the stock side. That decision comes during times of reallocation, as I don't really perseverate over the 70% increasing to 75 or dropping to 65 as that's the result of a moving market in either direction. Of course if one has individual stocks and not just a broad index, those stocks are monitored for potential sale. To make it simple, if your allocation were so aggressive you have no bond/cash position, you'd revert to the advice of selling the stock that's not one you'd buy today, one that no longer looks desirable.

  • It's a great point that it's perfectly OK to keep rainy day funds as stocks - they can be cashed out almost instantly.
    – Fattie
    Commented Apr 6, 2021 at 11:15
  • 2
    ? I said the opposite. That even the non-rainy day is not usually invested 100% in stock. Never suggested that the emergency/rainy-day be anything but cash/MM/CDs. Commented Apr 6, 2021 at 11:36

There is no definitive answer to this question which is why you are having trouble finding an answer with your search. A lot depends on your goals and investment style.

If you use an asset allocation model, which may include low cost index funds, you would sell those investments to keep your desired model. If your funds are out of balance, withdrawal could use to bring them back in line. Or, you would sell a small amount from each "bucket".

If you are a more active trader, seeking to maximize profits, you would sell those investments in which you deem less likely to be profitable in the future. How you determine that depends a lot upon how you determine your investments in the first place.

I recently had to do something similar myself. Since I am a asset allocation guy, I just sold off funds that preserved my asset allocation.

A good question to ask yourself about anything you own that could be sold is "Would I pay X to own that item today"? The X is what it could be sold for. If the answer is "no" then sell the thing.


Step 1: Rebalance your portfolio with the cash subtracted

Rebalancing is something you SHOULD be doing once a year or so anyway. If you don't: insert Step 0: learn how to rebalance and do it regularly Subtract the desired cash-out and calculate how much you want to have in each asset class after you are done. Than you can calculate the difference between the target amount and what you currently have: that will give you the amounts to withdraw and/or invest for each asset class.

Step 2: Look at the tax implications

This can be complicated and is highly depending on your local tax laws. Example: In the US long term gains are taxed less than regular income but withdrawing from retirement accounts can be expensive. Figure out which accounts and specific holdings are tax friendly. In some cases, you may have to adjust your allocation target somewhat. That's not a problem: regular rebalancing will eventually get your there over time.

Step 3: Look at past performance

but don't over-rotate on it and don't try to "predict" where things are going. There are different school of thoughts on this but a reasonably careful approach would be "preferentially sell the stuff that has been doing well" . Most asset classes go through cycles: if you sell a strong asset you cash in the gains while given the weaker ones more time to get to the strong part of their cycle. Don't sweat it too much though: you will always win some and loose some and as long as you are well diversified, that should all average out.

Step 4: Execute

Once you have figured what and how much to sell, withdraw and re-invest, you just need to do it. It helps enormously to have a "good" brokerage account and investments that are easily movable. For example I have a brokerage account at Vanguard (no advertisement intended, but I can recommend it) holding a very diverse set of ETFs. I could liquidate the whole thing in about 10 minutes. No fees, no delays in the execution of the trades, no brokers to call, etc. Don't overthink the timing: today will be better than tomorrow for some funds and worse for others. You have no way of predicting this, so you might as well just get it over with.


Buying a car ...

Consider the list "threw away 10,000 on strippers", "gambled away 10,000 at the casino", "spent 10,000 on a car".

The first two are fine: (A) they are one-time expenses and (B) at least you have a good memory. But "spent 10,000 on a car" is financial insanity. Because the costs are ongoing, monthly, extremely substantial, and never-ending.

Buying a car is financial madness.

Selling stocks to buy a car is a really unfortunate idea.

It's hard to imagine anything more negative than selling stocks to buy a car.

Buying a house...

Money put in to buying a car is pretty much the single most sickening waste of money possible. In contrast money put in to buying a house is, in almost all cases, an entirely good thing. So, great.

Of a group of stocks, which to sell: tax implications

Regarding the question "of a group of stocks, which to sell", in many jurisdictions there are tax differences between your stocks A, B, C, D due to issues such as length of time held.

If this is the case, you have the answer.

Of a group of stocks, which to sell: trading decision

Unfortunately, deciding which to sell of A, B, C, D is exactly the same as deciding which stock to buy of a group of stocks.

It is a straight-up stock picking exercise.

You literally have to time travel to the future and decide which one will go-up-least in the future.

As you have noted,

"Non Answers - I googled around and there seems to be a plethora of advice..."

Any advice on which stock to sell - or buy - from a group A, B, C, D is: bullshit. Total bullshit.

The final, unequivocal answer is: you must guess which one will go-up-least in the future, and that's it. Any other advice is as valueless as stock-tip newsletters unfortunately.

Regarding measures such as "volatility". If "volatility", or anything else, was a predictor of the future price of a stock - everyone would be a trillionaire. Nothing, at all, has predictive power of the future price of a stock; there is no stock picking.

One possible logical approach

People often ask "which stock to buy?" The only answer is "buy an index fund, because, obviously, it's completely impossible to stock pick."

Similarly, some would say that: if you have to cash out from stocks A, B, C, D. The only possible way to do this is to tell the future and know which one will go-up-least and sell that one. Hence, since stock picking is comic, the best thing to do is "sell the basket" .. equally sell from A, B, C, D.

In this way you minimize the risk of losses.

Speaking of an S&P index...

If one currently owns stocks A, B, C, D. It's better to sell them all and buy a plain broad index fund. One could take this opportunity to dump all of A B C D, take away the relevant cash for the house, strippers or gambling (forget the car: never sell stocks for a money-waste), and put the rest of the money in an index fund.

  • 4
    One never throws away money on strippers. That's an investment. In one's immediate happiness. A stock can rise or or fall, but the memory is with you forever..... Commented Apr 6, 2021 at 11:38
  • Quite right! And they do not have monthly insurance, repairs and consumables.
    – Fattie
    Commented Apr 6, 2021 at 11:56
  • @JTP-ApologisetoMonica I assumed that joke would end with a syphilis reference.
    – Stian
    Commented Apr 6, 2021 at 12:12
  • @Fattie How is buying a car universally a bad financial decision? Tons of people need cars to get to work (a fairly important thing for most people), and in a lot of cases buying a slightly better used car is more economically wise than buying a super cheap car. In my case I have bought two cars in the last 6ish years, one was $2200 and the other was $6000. In the years since we ultimately spent enough on repairs for the cheaper one that it was the more expensive car in the long run.
    – Kevin
    Commented Jul 21, 2021 at 22:39

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .