# Opening and closing balances when tracking a house value **URL:** \ **Category:** General\ **Tags:** hledger\ **Created:** [January 8, 2026, 7:21am UTC](https://forum.plaintextaccounting.org/t/opening-and-closing-balances-when-tracking-a-house-value/711 "2026-01-08T07:21:10Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![someguy](https://avatars.discourse-cdn.com/v4/letter/s/ac8455/32.png) [@someguy](https://forum.plaintextaccounting.org/u/someguy)\ **Post date:** [January 8, 2026, 7:21am UTC](https://forum.plaintextaccounting.org/t/opening-and-closing-balances-when-tracking-a-house-value/711/1 "2026-01-08T07:21:10Z")
Hi! I have been using hledger to track our personal expenses and things have been great so far. It has been slight over a year and I am still learning so much about accounting and the possibilities with hledger. As we move into 2026, I tried closing out 2025 and starting with a fresh file for 2026. Something I have not been able to comprehend has to do with tracking my house value in my journal. For example, lets say that this is my 2025 journal: ```auto 2025-01-01 Opening ; Bought a house years ago assets:house 1000000 USD @@ 800000 USD ; Bought for 800k, now worth 1m equity:opening 2025-06-30 Value update assets:house =1100000 USD ; house value went up again equity:unrealizedgains ``` Things are looking good at the end of the year: ```auto hledger -f example.journal bs Balance Sheet 2025-06-30 || 2025-06-30 ==============++============= Assets || --------------++------------- assets:house || 1100000 USD --------------++------------- || 1100000 USD ==============++============= Liabilities || --------------++------------- --------------++------------- || 0 ==============++============= Net: || 1100000 USD ``` Now it’s time to close 2025 and move this into 2026: ```auto hledger -f example.journal close --open --show-costs >example-2026-open.journal cat example-2026-open.journal 2026-01-08 opening balances ; clopen: assets:house 100000 USD assets:house 1000000 USD @@ 800000 USD = 1100000 USD equity:opening/closing balances ``` My first surprise is that I now have 2 different assets/accounts for the house. Reading the documentation I think this might have to do with lots but I am unsure. I would have expected one line/account which would have stated I have an assets of 1,1m which has an acquisition cost of 800k… Looking at the situation from 2026, I see this: ```auto hledger -f example-2026-open.journal bs Balance Sheet 2026-01-08 || 2026-01-08 ==============++============= Assets || --------------++------------- assets:house || 1100000 USD --------------++------------- || 1100000 USD ==============++============= Liabilities || --------------++------------- --------------++------------- || 0 ==============++============= Net: || 1100000 USD ``` Which is what I expect, but, if I look at the gains, I see: ```auto hledger -f example-2026-open.journal bs --gain Balance Sheet 2026-01-08 (Historical Gain), valued at period ends || 2026-01-08 ==============++============ Assets || --------------++------------ assets:house || 200000 USD --------------++------------ || 200000 USD ==============++============ Liabilities || --------------++------------ --------------++------------ || 0 ==============++============ Net: || 200000 USD ``` The gain is “only” 200k, not the 300k I would have expected (1,1m value - 800k acquisition cost). What am I doing wrong? ---
**Author:** ![simonmic](https://yyz1.discourse-cdn.com/flex003/user_avatar/forum.plaintextaccounting.org/simonmic/32/1_2.png) [@simonmic](https://forum.plaintextaccounting.org/u/simonmic)\ **Post date:** [January 8, 2026, 8:17am UTC](https://forum.plaintextaccounting.org/t/opening-and-closing-balances-when-tracking-a-house-value/711/2 "2026-01-08T08:17:40Z")
> My first surprise is that I now have 2 different assets/accounts for the house. It's because of the different costs: ```auto assets:house 1000000 USD @@ 800000 USD ; Bought for 800k, now worth 1m ``` posts 1000000 USD with a total cost of 800000, and ```auto assets:house =1100000 USD ; house value went up again ``` posts 100000 USD with no cost. These amounts are handled almost like two separate commodities. Reports that don't show costs will show them summed together, but `close --show-costs` shows them separately. > The gain is “only” 200k, not the 300k I would have expected I'm not sure what the explanation is for this yet. I haven't used the --gain report much. And, I'm not sure that journal entry makes sense; I have not seen the same commodity used on both sides of @ or @@ before. I usually see it done like this: ```auto 2025-01-01 Opening assets:house 1 HOUSE @@ 800000 USD ; Bought this house years ago for 800k equity:opening ; Value updates P 2025-01-01 HOUSE 1000000 USD P 2025-06-30 HOUSE 1100000 USD ``` Then `bs -B` shows the cost, `bs -V [-e DATE]` shows the value on DATE, and `bs --gain [-e DATE]` shows the gain/loss as of DATE. ---
**Author:** ![someguy](https://avatars.discourse-cdn.com/v4/letter/s/ac8455/32.png) [@someguy](https://forum.plaintextaccounting.org/u/someguy)\ **Post date:** [January 8, 2026, 9:27pm UTC](https://forum.plaintextaccounting.org/t/opening-and-closing-balances-when-tracking-a-house-value/711/3 "2026-01-08T21:27:34Z")
Thank you for the quick reply! > [@simonmic](#): > > Reports that don't show costs will show them summed together, > but `close --show-costs` shows them separately. This was insightful!! > [@simonmic](#): > > And, I'm not sure that journal entry makes sense; You are of course right,. and then you also said: > [@simonmic](#): > > I'm not sure what the explanation is for this yet. Which lead me to further play around and that produced and even stranger result (notice the 06-30 statement which includes both value and cost - albeit using the same currency on both side of the @@): ```auto % cat example2.journal 2025-01-01 Opening ; Bought a house years ago assets:house 1000000 USD @@ 800000 USD ; Bought for 800k, now worth 1m equity:opening 2025-06-30 Value update assets:house =1100000 USD @@ 800000 USD ; house value went up again equity:unrealizedgains % hledger -f example2.journal bs --gain Balance Sheet 2025-06-30 (Historical Gain), valued at period ends || 2025-06-30 ==============++============ Assets || --------------++------------ assets:house || 500000 USD --------------++------------ || 500000 USD ==============++============ Liabilities || --------------++------------ --------------++------------ || 0 ==============++============ Net: || 500000 USD ``` Which is too many an even stranger number… likely a bug somewhere, or perhaps a statement which should be rejected? I then tried your proposed approach: ```auto 2025-01-01 Opening ; Bought a house years ago assets:house 1 HOUSE @@ 800000 USD ; Bought for 800k, now worth 1m equity:opening P 2025-01-01 HOUSE 1000000 USD P 2025-06-30 HOUSE 1100000 USD ; House value went up P 2025-09-30 HOUSE 900000 USD ; And house value went down ``` And this appears to work with my use cases. ---
**Author:** ![adept](https://yyz1.discourse-cdn.com/flex003/user_avatar/forum.plaintextaccounting.org/adept/32/36_2.png) [@adept](https://forum.plaintextaccounting.org/u/adept)\ **Post date:** [January 9, 2026, 12:27am UTC](https://forum.plaintextaccounting.org/t/opening-and-closing-balances-when-tracking-a-house-value/711/4 "2026-01-09T00:27:16Z")
My unsolicited opinion is that USD @@ USD should be treated like an error and rejected. So many weird implications otherwise.. For example, if you try to –Infer-market-prices from this and then –value the balances in USD, should they be inflated/discounted because of this transaction? 🙂 ---
**Author:** ![simonmic](https://yyz1.discourse-cdn.com/flex003/user_avatar/forum.plaintextaccounting.org/simonmic/32/1_2.png) [@simonmic](https://forum.plaintextaccounting.org/u/simonmic)\ **Post date:** [January 9, 2026, 8:02am UTC](https://forum.plaintextaccounting.org/t/opening-and-closing-balances-when-tracking-a-house-value/711/5 "2026-01-09T08:02:41Z")
I'm inclined to agree, it's fun to imagine what meaning we could find for it but nothing useful immediately comes to mind. Not that it has been causing much trouble to date.