|
oliveoil posted:* - How many small-medium cities in the US have gone from being decent, normal places to complete collapse within 15 years? And that's the worst-case scenario. You've basically described a lot of the rust belt and what happened to them from the '70s to '90s.
|
#
?
Jan 25, 2017 02:20
|
|
|
|
| # ? May 10, 2024 13:37 |
|
|
People still live there and pay rent. Rent is lower than most places, but it's not gone and on a triplex it'll still almost certainly be enough to pay back your loan, leaving you with a triplex that cost $15k. Even if it's only in an area that most people left, you'd have to have like, Detroit levels of failure for a $15k triplex to be a bad investment.
|
|
#
?
Jan 25, 2017 03:01
|
|
|
My take is that people often under estimate the potential time requirement for such an endeavor. That and maintenance costs can escalate quickly. Getting in on a good deal with well understood maintenance costs is key to success. Many landlords to be don't have the expertise to due the required diligence.
|
|
#
?
Jan 25, 2017 04:39
|
|
|
Oh, and I read that even though I've maxed my 401k contribution, I can still make a deductible contribution to my IRA for 2016. If I do that before I file my taxes, will I get the deductible for 2016 or have to wait till next year to see it?
|
|
#
?
Jan 25, 2017 05:42
|
|
|
oliveoil posted:Oh, and I read that even though I've maxed my 401k contribution, I can still make a deductible contribution to my IRA for 2016. If I do that before I file my taxes, will I get the deductible for 2016 or have to wait till next year to see it? If you do it before your 2016 taxes, you get to claim it on your 2016 taxes. But be aware of the traditional IRA deduction limits based on income. If you're in a high enough income situation where you're maxing your 401k, you may likely be disqualified from deducting traditional IRA contributions. Refer to the IRS site here: https://www.irs.gov/retirement-plans/ira-deduction-limits Roth IRA contributions aren't deductible so they don't affect your taxes either way.
|
|
#
?
Jan 25, 2017 06:29
|
|
|
oliveoil posted:Can you really? Can't you buy a $300k triplex with 5% down? Then you collect the full benefit of a 6% annual return on $300k... I.e., $18k/yr. If you spend that 5% down-payment - $15k - on shares of some REIT(s), you're going to get 6% on your $15k, which is like $900/yr. This is laughably wrong for one simple reason: you have to pay interest on the loan in addition to securing some amount of income on your investment in the triplex. If you bought a $300k triplex with 5% down you are taking a loan for $285,000 out. What kind of loan do you get? If you can get a 6% return per annum, your monthly income is around $1500. A 30-year fixed at 4.125% means you're paying $1381.25 just to cover PI, and you still have to tack taxes & insurance on top. If you could pay no tax, and pay no insurance, you're already down to $1425/year cash flow vs the $900/year dividends for the REIT. Now, sure, you are building equity in the property, but having bought one specific property you are not diversified. What if the market you buy in crashes hard? What if you get some tenants who destroy the interior of the home? Can you maintain the 3 units in the home to the point of preserving the value of the triplex on $525 / year? If you spend any more than that on maintenance you'd have come out ahead with the REIT. There's risk in both, but the risk profile of owning the individual property in whole is WAY higher for not much more additional gain. I'd toss the $15k into the REIT. Oh, did we talk about closing costs? Closing costs...
|
|
#
?
Jan 25, 2017 15:08
|
|
|
But $1381 toward principal and insurance means I'm building equity at a rate of $1381 * 12 = $16,572/year, which is way higher than the $900/yr I'd get from an REIT. The REIT's cash flow may be slightly better but the rate of increase in my net worth from the REIT is $900/16,572 = 5.4% of what the triplex would give me. That is, the triplex gives me a return almost 20x better.
|
|
#
?
Jan 25, 2017 16:44
|
|
|
oliveoil posted:But $1381 toward principal and insurance means I'm building equity at a rate of $1381 * 12 = $16,572/year, which is way higher than the $900/yr I'd get from an REIT. The REIT's cash flow may be slightly better but the rate of increase in my net worth from the REIT is $900/16,572 = 5.4% of what the triplex would give me. That is, the triplex gives me a return almost 20x better. This is a good troll, I give it an 8.5.
|
|
#
?
Jan 25, 2017 16:48
|
|
|
Actually, I just had a thought. Isn't 5% down on a $300k triplex basically like taking out a $285k loan and putting it all in REITs? Because that actually sounds like a better idea. Basically even safer leverage.
|
|
#
?
Jan 25, 2017 16:49
|
|
|
Yes, buying a home is the only way most people interact with leveraged investments, which can make it very lucrative. It can also ruin you personally and destroy civilization, so there's that.
Pryor on Fire fucked around with this message at 17:00 on Jan 25, 2017 |
|
#
?
Jan 25, 2017 16:58
|
|
|
oliveoil posted:Actually, I just had a thought. Isn't 5% down on a $300k triplex basically like taking out a $285k loan and putting it all in REITs? Because that actually sounds like a better idea. Basically even safer leverage. No, REITs themselves are leveraged. It is more like putting your down payment (i.e. $15k) in REITs, although it depends on the REIT. Some are more modestly leveraged than others. Also, you should consider the cost of hiring a management company for your triplex when comparing it with REITs. 80k fucked around with this message at 17:08 on Jan 25, 2017 |
|
#
?
Jan 25, 2017 17:03
|
|
|
80k posted:No, REITs themselves are leveraged. Sorry, I don't understand. Do you mean that REITs use their funds to essentially take out loans on many properties with like 20% down, essentially getting the same leverage I'd have if I already had a first home and was looking to buy a second property as an investment? If that's the case, as an investor in REITs, how does the value of increasing equity in these properties get returned to me? I.e., if we considered a simple REIT with a single property purchased with 20% down... What happens as the loan is paid off? Does the value of my share in the REIT increase? Does this mean that if I want more leverage, I should look for RWITs that are actively acquiring properties / invested in many properties where they don't yet have their loans fully paid off?
|
|
#
?
Jan 25, 2017 17:44
|
|
|
oliveoil posted:But $1381 toward principal and insurance means I'm building equity at a rate of $1381 * 12 = $16,572/year, which is way higher than the $900/yr I'd get from an REIT. The REIT's cash flow may be slightly better but the rate of increase in my net worth from the REIT is $900/16,572 = 5.4% of what the triplex would give me. That is, the triplex gives me a return almost 20x better. I... I don't even.... I would say this is a troll, except what self aware troll would post something like this in the FI thread, when there are way better threads to try and get away with it. oliveoil posted:Sorry, I don't understand. Clearly. If somehow this isn't a troll, we would be happy to point you to a more appropriate thread that can help you down the path of For my own FI goals, I'm planning on moving out of So. Cal to somewhere way more affordable to help me in my pursuits of FI. Best part is my current job will let me work remotely. Currently looking at Idaho. 10-18 month timeline on the move, depending on when I knock my wife up. Her taking maternity leave and then moving is our ideal situation right now. Being born and raised in one of the most expensive places on earth is BWM. Granted, if I end up in Idaho, my kids will probably move to California for college, so the cycle will continue. Loan Dusty Road fucked around with this message at 08:27 on Feb 11, 2017 |
|
#
?
Feb 11, 2017 08:18
|
|
|
I'm in Australia if it matters. I bank with commonwealth bank and have bought shares in the past through commsec (owned by the bank). I also have my superannuation with a company owned by commbank. This is nice because i can log into online banking and see absolutely everything. I can transfer cash into the linked money market account and then buy shares at $19.95 a trade. I'm going to start investing in Vanguard instead of other stupid things but I'm confused about the actual way to invest with Vanguard. Assuming i'm going to be putting $1,500 every two weeks toward Vangaurd, do i buy ETFs at $19.95 a trade? If i do this monthly it effectively adds 0.67% to the cost (i think? 3000/19.95=0.00665) , but its a one-off cost. The ETFs are 0.14% expense ratio. detail is here https://www.vanguardinvestments.com.au/retail/ret/investments/product.html#/fundDetail/etf/portId=8205/?overview or do i buy into the retail fund tracking the exact same thing. assuming the 5k minimum buy-in is fine. then i can put money in with Bpay transfer (slightly easier than doing the trade myself once a month). costs are 0.75% on the first 50,000$, 0.50% on the next 50,000$ and 0.35% on balance over 100,000$. The Buy/Sell spread cost is 0.10%/0.10%. detail is here https://www.vanguardinvestments.com.au/retail/ret/investments/product.html#/fundDetail/retail/portId=8129/?overview a) i dont know what buy sell spread cost is b) i dont know how to compare the cost i would be buying off the market at (assuming there is even enough liquidity in the ETFs - it seems there is though) to the cost the fund uses for each unit c) i can't work this out apples vs apples d) does it matter than the etf distributes dividends quarterly but the fund is semi annual? dividend reinvestment seems to be available on both ETFs and the fund
|
|
#
?
Feb 11, 2017 08:19
|
|
|
Made these for my own use, but others might find them interesting: The model is that you start with some amount savings, and each year
|
|
#
?
Feb 11, 2017 10:02
|
|
|
Always worth it to do a monte carlo on these sorts of things, it's not a distribution with finite volatility (even tho of course in empirics the volatility is finite). Fat tailled distribution, not Gaussian Many people do monte carlo on US volatility only. You should also do monte carlo on "communists or equivalent take over and expropriate all private property you bourgeoisie scum" risk, which remains a nonzero chance in the US. Monte carlo on securities from 1880 on in Russian securities is a bit miserable in the 1920-1980 period there
|
|
#
?
Feb 11, 2017 16:50
|
|
|
curufinor posted:
This is often mentioned as an advantage to having some international exposure, but I think it works in the reverse. Your international stocks held in an account at a US bank are going to be forfeit regardless. In fact, you are open to political risk in your own country AND the foreign country. To the simulation point: all available evidence points to a T distribution as the best model for total stock (or bond) returns on a yearly or monthly basis, with zero autocorrelation, expected return = GDP growth + buybacks + dividends (so about 6.5% for US), and stddev set to historical (you can go a little higher to account for parameter uncertainty, or implement actual parameter uncertainty if you're that spergy). For REITs I'd go with expected return = historical but adjusted for the fact that a large part of real estate return in the time the class has existed was from falling interest rates (not an expectation going forward). For bonds, use the fund's YTM (again, previous returns are artificially high from falling rates) or some kind of calculation based on the current yield curve and the types of durations the fund buys/sells at. Also, if you're going to do a full-blown sim with asset allocations it's worth pointing out that you should get the correlations between asset classes right (hint: all the major classes have positive correlation over annual+ time spans)
|
|
#
?
Feb 11, 2017 19:49
|
|
|
Eyes Only posted:(hint: all the major classes have positive correlation over annual+ time spans) Can you please explain what you mean by this or give an example?
|
|
#
?
Feb 11, 2017 20:03
|
|
|
oliveoil posted:Can you please explain what you mean by this or give an example? People commonly try to use observed daily correlations between asset classes as an indicator of future correlation. The usual reason is "correlations change" which makes sense if you're running a day to day portfolio, but for long term investment it falls apart. It's just how time series works; even with constant correlation, short term correlations are dampened over longer time periods. With "changing" correlation the short term empirical results are even less useful as they lose credibility as a result of their transience. As an example, there are a lot of people out there floating assumptions for US stock:bond correlation at -20%. Wealthfront publishes their matrix and they have US govt bonds around there vs US stock. To put this in perspective, the empirical annual correlation for (two easy to find pieces of data) S&P500:BarclaysAggBond is +22.5% since 1976. It is worth noting that the 90% confidence interval for annual correlation over 41 years is +/-24% based on the standalone variances of those two individual classes - the lower bound just barely goes under 0. It is really hard to justify negative correlation here, and that huge interval really outlines how much any observed short term correlation is simply caused by the underlying variances of the two assets. This is super important to get right and a lot of people with impressive-sounding credentials get this stuff wrong to disastrous effect. For examples of horrible mistakes see bank collapses in the GFC and Nate Silver vs the world in modeling last year's election. In my line of work (P&C insurance modeling) the only lines of business I've seen with negative entries on the matrix have either been statistically insignificant or have had extreme drawbacks (lower returns, higher risk) to make up for their lower/negative correl. It makes me seriously raise an eyebrow at any such assumption anywhere in finance. I can agree with "correlations change" on a qualitative level - it's easy to imagine the world being more connected today than in 1976 so US-Intl should be more tightly coupled than in the past, but that doesn't exactly hurt my argument against negatives.
|
|
#
?
Feb 11, 2017 22:24
|
|
|
Eyes Only posted:To the simulation point: all available evidence points to a T distribution as the best model for total stock (or bond) returns on a yearly or monthly basis Do you mean log-t? Or do you just put a t distribution directly on the returns and throw away any nonsense samples? Most of this lines up with my intuitions; I do multivariate Gaussian time series on non-financial data, so it's useful to see somebody else's modeling assumptions.
|
|
#
?
Feb 12, 2017 04:19
|
|
|
Dwight Eisenhower posted:This is laughably wrong for one simple reason: you have to pay interest on the loan in addition to securing some amount of income on your investment in the triplex. He's worded so many things wrong here, but you're not factoring in him living in one of the units. I have no idea what he means by "return" since you don't base your return off the full purchase price, and a 6% cap rate would be way too high for a triplex with one unit not collecting. But let's say each unit rents for $750 just for fun. That gives us the $1,500 a month in income we want, then we take out the mortgage of $1381.25, and let's just say $400 a month in additional expenses. So that leaves a -$281.25, but he's living in a unit rent free that goes for $750. So he's really benefitting by $468.75 a month, or $5,625 a year. His return is based on what money he put in (the $15,000), so it's a 37.5% return. Now, if all 3 units rent for $1,500 a month, clearly it was a poo poo deal. REITS and rental properties are nowhere even close to comparable and I don't understand why people keep acting like they are.
|
|
#
?
Feb 12, 2017 04:35
|
|
|
Alhireth-Hotep posted:Do you mean log-t? Or do you just put a t distribution directly on the returns and throw away any nonsense samples? Most of this lines up with my intuitions; I do multivariate Gaussian time series on non-financial data, so it's useful to see somebody else's modeling assumptions. Just a t directly on the returns, 3 DoF fits best for most classes, then re-center and scale as needed. Redraw any samples <= -1 (and >= 1 if you care to maintain symmetry). You will need to tweak your parameters since the redraws will lower the variance (and raise the EV if you broke symmetry).
|
|
#
?
Feb 12, 2017 06:19
|
|
|
Jon Von Anchovi posted:I'm in Australia if it matters. I bank with commonwealth bank and have bought shares in the past through commsec (owned by the bank). I also have my superannuation with a company owned by commbank. This is nice because i can log into online banking and see absolutely everything. I can transfer cash into the linked money market account and then buy shares at $19.95 a trade. Hey mate, I can't answer all your questions but I can tell you the decisions I made for myself. I had an initial lump sum (30k+) so I used a NaBtrade '30 free trades on signup' to bulk-buy ETFs with no trading costs. No brainer. Then saved up the 5k for the retail fund, at a similar rate to yourself. Then, when the balance for that fund reached 100k, I switched to a wholesale fund of the same type. (The website says $500k - call them up. It's actually 100.) For me the benefits of the easy regular BPAY and auto-rebalancing/re-investing for the funds far outweighs the sub 0.5% difference in fees. (Particularly for the wholesale funds the difference is minimal) Also minor point but some of the ETFs do not have reinvestment options (eg VTS/VEU) and every quarter I have to juggle annoying little dividends from my ETFs. Bottom line is there is not a huge difference and it's basically convenience (Funds) vs best possible long-term return with extra work involved and short-time higher costs (ETFs)
|
|
#
?
Feb 12, 2017 07:52
|
|
|
curufinor posted:Always worth it to do a monte carlo on these sorts of things, it's not a distribution with finite volatility (even tho of course in empirics the volatility is finite). Fat tailled distribution, not Gaussian I think estimating those kind of tail risks is more suited to apocalypse planning than retirement planning. In the event, you'd have bigger problems than not being able to live off your investments. Eyes Only posted:Just a t directly on the returns, 3 DoF fits best for most classes, then re-center and scale as needed. Redraw any samples <= -1 (and >= 1 if you care to maintain symmetry). You will need to tweak your parameters since the redraws will lower the variance (and raise the EV if you broke symmetry). coffeetable fucked around with this message at 10:53 on Feb 12, 2017 |
|
#
?
Feb 12, 2017 09:13
|
|
|
coffeetable posted:Where does this sampling scheme come from? When modeling returns directly you have to limit them to a lower bound of -100% to avoid nonsense results. The two common ways to do that are by capping at the bound, or resampling any draws that fall out of it. The former introduces a mass point at -100% (or -99.9%, -99.9999% etc), desirable for individual stocks/bonds. The latter neatly makes the sample density approach zero as you near -1, just like a log distribution would do, more appropriate for indices/funds. Both methods will alter your variance (slightly; the density out there is pretty small in practice) so you need to adjust for that. The reason I advocate this over directly modeling price via a log distribution is it gives you the flexibility above, plus it generally simplifies the logic of the rest of the simulation.
|
|
#
?
Feb 12, 2017 19:42
|
|
|
Eyes Only posted:When modeling returns directly you have to limit them to a lower bound of -100% to avoid nonsense results. The two common ways to do that are by capping at the bound, or resampling any draws that fall out of it. The former introduces a mass point at -100% (or -99.9%, -99.9999% etc), desirable for individual stocks/bonds. The latter neatly makes the sample density approach zero as you near -1, just like a log distribution would do, more appropriate for indices/funds.
|
|
#
?
Feb 12, 2017 19:59
|
|
|
Chadzok posted:Hey mate, I can't answer all your questions but I can tell you the decisions I made for myself. Thanks mate - this was super useful. I was comparing multiple ETFs with multiple funds but didn't consider rebalancing; that comment made me go looking further and realise I could go for Vanguard high growth index fund and it has the 90/10 stocks bonds allocation I'd probably be going for. I think I was thinking too hard about The broker cost and buy sell spreads. They are a one time cost and tiny in the long run. I'm still tossing up between just buying VAS and VTS and going to high growth retail fund, especially since I hadn't considered switching to wholesale was an option.
|
|
#
?
Feb 12, 2017 22:22
|
|
|
Whoops, Google paid a bunch of engineers working on self-driving cars too much, and they left because after some bonuses they really didn't need to work anymore: One Reason Staffers Quit Google's Car Project? The Company Paid Them So Much quote:'F-you money' awarded to veteran team members boosted parent Alphabet’s R&D costs, prompting comment from the company's CFO SpelledBackwards fucked around with this message at 10:16 on Feb 14, 2017 |
|
#
?
Feb 14, 2017 10:11
|
|
|
Eyes Only posted:simulation stuff Couple things - 1) if you want to use the gordon model for forecasting equity expected returns, most academics/practitioners are forecasting significantly lower returns - on the order of 3-4% compared to your 6.5%. GDP growth has actually been a ceiling, not the floor for dividend growth with the latter tending to capture half of GDP growth. It's not surprising when you consider public equities not capturing entrepreneurial growth very well. 2) on MVO, even wealthfront kinda shrugs and falls back on black-litterman to back out expected returns. stuff like this makes me skeptical about roboadvisors and their value add, any random dude can run an MVO with a bunch of numbers and construct an 'optimized' portfolio. garbage in, garbage out.
|
|
#
?
Feb 14, 2017 11:32
|
|
|
SpelledBackwards posted:Whoops, Google paid a bunch of engineers working on self-driving cars too much, and they left because after some bonuses they really didn't need to work anymore: I like how two of them referred to it as gently caress you money. It's interesting that more people are starting to see that there's more to life than clocking up money like a high score. Your net worth isn't a video game.
|
|
#
?
Feb 15, 2017 00:23
|
|
|
Devian666 posted:I like how two of them referred to it as gently caress you money. It's interesting that more people are starting to see that there's more to life than clocking up money like a high score. Your net worth isn't a video game. It will be interesting to see the interplay between highly skilled engineers peacing out and engineering salaries. A money spiral?
|
|
#
?
Feb 15, 2017 00:29
|
|
|
baquerd posted:It will be interesting to see the interplay between highly skilled engineers peacing out and engineering salaries. A money spiral? One of the things that always comes up with FI is doing things that you like doing. I still like the engineering work that I do and run my own business. My retirement fund is increasing in size at a rapid pace but will that affect my will to carry on doing engineering? Unlikely. It would take something pretty major for me to want to shut down everything and peace out. I work with a number of other engineers that have no financial need to carry on working.
|
|
#
?
Feb 15, 2017 02:31
|
|
|
shrike82 posted:1) if you want to use the gordon model for forecasting equity expected returns, most academics/practitioners are forecasting significantly lower returns - on the order of 3-4% compared to your 6.5%. GDP growth has actually been a ceiling, not the floor for dividend growth with the latter tending to capture half of GDP growth. It's not surprising when you consider public equities not capturing entrepreneurial growth very well. Capital in the 21st Century is a good read for anyone interested in financial independence. Good perspective on some long term macro trends that tend to be overlooked by most analysis.
|
|
#
?
Feb 15, 2017 03:24
|
|
|
AreWeDrunkYet posted:Capital in the 21st Century is a good read for anyone interested in financial independence. Good perspective on some long term macro trends that tend to be overlooked by most analysis. That's quite a lengthy read. My capital gains and income from investments are greatly exceeding GDP so I'm doing my part to increase inequality.
|
|
#
?
Feb 15, 2017 03:39
|
|
|
Devian666 posted:That's quite a lengthy read. My capital gains and income from investments are greatly exceeding GDP so I'm doing my part to increase inequality. I hope you mean GDP per capita, otherwise no complaining about your retaining wall when you could be living in a space station instead.
|
|
#
?
Feb 15, 2017 03:53
|
|
|
SpelledBackwards posted:I hope you mean GDP per capita, otherwise no complaining about your retaining wall when you could be living in a space station instead. I just assume that's implied. You'd know if I was collecting the GDP of a nation as Trump's adviser would have been on the phone to my space station instead of Putin.
|
|
#
?
Feb 15, 2017 04:50
|
|
|
So I'm thinking about leaving my current job which has a stupid good 401k - 7% of base pay automatic match (not even a match? They said even if I contributed 0 they would always put in 7%, its kind of crazy) for the reasons why people leave jobs - terrible management, crazy work load, need a change of scenery, etc. So I just interviewed at another company that seems pretty cool except I looked at their 401k info and they said that max contributions to 25% of pay. That won't get me to my 18k per year unfortunately. Has anyone ever seen this before? To keep our FI goals on track, we can rebalance some other stuff but it bothers me that this is a thing. I was trying to figure out what thread to post this question in and I guess this one seemed the most appropriate since it mostly interferes with getting FI faster.
|
|
#
?
Mar 15, 2017 15:08
|
|
|
Have you negotiated salary yet? It's kinda premature to be thinking about 401k matching if a higher salary would make it moot. Even if you don't make 72k, it's totally moot if your salary goes up by, say, 10% by changing companies. Are you maxing out an IRA as well? Trust me your job satisfaction and base pay both matter more than your 401k matching percentage. Putting more than 25% of my earnings into a 401k is a pretty frightening thought to me - 65 is a long time from now.
|
|
#
?
Mar 15, 2017 15:38
|
|
|
silicone thrills posted:So I just interviewed at another company that seems pretty cool except I looked at their 401k info and they said that max contributions to 25% of pay. That won't get me to my 18k per year unfortunately. Did you know employer match has nothing to do with your personal max contribution? You can still put in $18k of your own money (assuming you haven't hit the combined max contribution limit that is something like $53k if I recall).
|
|
#
?
Mar 15, 2017 15:56
|
|
|
|
| # ? May 10, 2024 13:37 |
|
|
Jeffrey of YOSPOS posted:Have you negotiated salary yet? It's kinda premature to be thinking about 401k matching if a higher salary would make it moot. Even if you don't make 72k, it's totally moot if your salary goes up by, say, 10% by changing companies. Are you maxing out an IRA as well? I'm the low earner in my relationship if that helps put it into perspective. I'm 29 and we have the potential to be FI in about 6-7 years if we keep on track. The plan is to try to negotiate up to 72 with this in mind for sure. There's enough jobs in my area that I can wait out for something better. Anyway, I was just curious if anyone else had ever seen 401k contributions capped like that. I've had alot of job offers in the past month and that's probably one of the odder things i've seen. Answer to the above - It specifically says 25% of base pay. It isn't the match i'm talking about.
|
|
#
?
Mar 15, 2017 15:59
|
|