All Forum Posts by: Bill F.

You started your original post with "If real estate is a stock..." weird now to do a 180 and state the obvious that they don't actually behave like stocks. 

Its axiomatic that they differ for a host of reasons, fungibility, transaction costs, settlement time, ect.  I thought we all understood these facts as table stakes. Shame on me for assuming. It won't happen again. Hate to burst your bubble about stocks, but even exchanges don't have one price. They can't with a CLOB. Neverminded trades that take place in ATS or through market market making firms w/ PFOF.

I never said it was the right way to think about it either lol; there can't be a right or wrong with a thought experiment, just the insights you take away.  I said I was curious to look at it through a quant lens. 

Question for you, would you still have the same opinion if the results yielded a Sharpe ratio of 5 and proved you right?  

Quote from @Carlos Ptriawan:

Quote from @Bill F.:

Quote from @Carlos Ptriawan:

Quote from @Bill F.:

@Carlos Ptriawan

You piqued my interest framing this in terms of a trading strategy so I looked at it though the quant lens. 

Interesting conclusion, as a trading strategy it doesn't fair so well, with Sharpe Ratio for the most expensive MSA's all below 1. 

I took the Case Shiller index data going back to 2000 and looked for patterns in highs and lows. By inspection I saw that Jan has the lowest prices and usually July has the highest. I figured out the MSA's with average prices over $500k as of 2023 and calculated the return if you bough in Jan and sold 6 months later in July or 18 months later, also in July. 

This doesn't take into account transaction costs and I didn't do anything beyond 18 months since that isn't really trading per se. 

Overall interesting idea to get a few extra points of yield if you are a buy and hold investor. 

 There's problem with case-shiller, it's lagging data. I would show it to you which indicator to use.

Fair enough, Case Shiller is lagged, but it is consistently lagged so while the absolute values won't be accurate, the relative magnitude of change remains the same.  If the values under estimate by a factor of 50%, then the modified Sharpe for a 18mo hold in San Francisco is 0.9. Not exactly blowing the doors off and making Jim Simmons shake in his boots. 

Never mind your best geometric return for an 18mo hold from 2000-2021 is 10% and it goes down to 9% once you factor in 1% transaction costs (which we all know is low). The Total Market Index did 7% over that same period. But the high StDev relative the average is what murders this as a trading strategy over the long run.  

The data you linked to doesn't go a granular enough MSA level so you can't limit the analysis to areas with homes > $500k. It also only goes back to 2016 and is collected monthly which makes the sample size= 78, far too small. 

I know this analysis is good from far, but far from good. Using a lagged and indexed data source isn't the best. Picking buying and selling date via graphical inspection only gets you so far... the list goes on. I'm doing this before work; this is the best I'll give strangers on the internet for free ;)

The slightly prettier back of the napkin math shows that while this is an interesting idea, it not some factory of untapped alpha. 

 I agree with that, you still need to adjust it to make it work to your specific market and still, the biggest risk in real estate is overbought.
So if you buy the wrong property you are still screwed .

For me the data from Fed is consistent with price behavior in our particular market, it's more deterministic when the interest rate is stable, not when it's volatile, but if interest rate is easing and seasonality factor is in, there's huge advantage in term of reward/risk play.

 I agree with you Carlos. The seasonality certainly exists due to the forces driven by market participants habits/desires. When markets are going up and to the right plus you have enough capital to limit the individual property risk, this works and works well. 

But when it goes wrong, it goes wrong fast. As with a lot of things, this relies on a degree of market timing. That fits in some ppl's wheelhouse and not for others. 

No wrong or right, just shades of different.

anyway,  interesting idea to look at. Thanks for brining this up! 

 Hi Erik, 

Curious what licenses and or certs your professionals have? Can you tell us that? 

Quote from @Carlos Ptriawan:

Quote from @Bill F.:

@Carlos Ptriawan

You piqued my interest framing this in terms of a trading strategy so I looked at it though the quant lens. 

Interesting conclusion, as a trading strategy it doesn't fair so well, with Sharpe Ratio for the most expensive MSA's all below 1. 

I took the Case Shiller index data going back to 2000 and looked for patterns in highs and lows. By inspection I saw that Jan has the lowest prices and usually July has the highest. I figured out the MSA's with average prices over $500k as of 2023 and calculated the return if you bough in Jan and sold 6 months later in July or 18 months later, also in July. 

This doesn't take into account transaction costs and I didn't do anything beyond 18 months since that isn't really trading per se. 

Overall interesting idea to get a few extra points of yield if you are a buy and hold investor. 

 There's problem with case-shiller, it's lagging data. I would show it to you which indicator to use.

Fair enough, Case Shiller is lagged, but it is consistently lagged so while the absolute values won't be accurate, the relative magnitude of change remains the same.  If the values under estimate by a factor of 50%, then the modified Sharpe for a 18mo hold in San Francisco is 0.9. Not exactly blowing the doors off and making Jim Simmons shake in his boots. 

Never mind your best geometric return for an 18mo hold from 2000-2021 is 10% and it goes down to 9% once you factor in 1% transaction costs (which we all know is low). The Total Market Index did 7% over that same period. But the high StDev relative the average is what murders this as a trading strategy over the long run.  

The data you linked to doesn't go a granular enough MSA level so you can't limit the analysis to areas with homes > $500k. It also only goes back to 2016 and is collected monthly which makes the sample size= 78, far too small. 

I know this analysis is good from far, but far from good. Using a lagged and indexed data source isn't the best. Picking buying and selling date via graphical inspection only gets you so far... the list goes on. I'm doing this before work; this is the best I'll give strangers on the internet for free ;)

The slightly prettier back of the napkin math shows that while this is an interesting idea, it not some factory of untapped alpha. 

@Carlos Ptriawan

You piqued my interest framing this in terms of a trading strategy so I looked at it though the quant lens. 

Interesting conclusion, as a trading strategy it doesn't fair so well, with Sharpe Ratio for the most expensive MSA's all below 1. 

I took the Case Shiller index data going back to 2000 and looked for patterns in highs and lows. By inspection I saw that Jan has the lowest prices and usually July has the highest. I figured out the MSA's with average prices over $500k as of 2023 and calculated the return if you bough in Jan and sold 6 months later in July or 18 months later, also in July. 

This doesn't take into account transaction costs and I didn't do anything beyond 18 months since that isn't really trading per se. 

Overall interesting idea to get a few extra points of yield if you are a buy and hold investor. 

 Interesting, thanks for the insight. Lets say the seller knows you are a wholesaler working with a list of buyers and has a concern about none of your buyers will have an interest in the property. They asks if you have the ability to close the deal personally? How do you handle that? 

 Have to disagree with your #1 Joe since it goes counter to the nature of costs in RE. I agree that OpEx expense growth will grow in parallel with rent grow, however it makes around half of the total expenses, with the other half being debt, which is fixed. So the cash flow will rise in a related way to the rent growth. 

Simple example to illustrate. I make widgets and sell them for $5. Their total cost is $4, $2 of raw materials and $2 in fixed payment loans. My Cash Flow is $1. If prices and costs go up 10%, now I sell the widget for $5.50 and raw materials cost me $2.20, but my loan hasn't changed and still costs $2, so my total cost is $4.20 and profit is $1.30. 

The only way for the profit to remains $1 if prices go up by 10% is for expenses to go up by 25%. That can happen, but that 25% increase in expenses usually isn't happening in new builds/ rehabbed homes, its buying stuff that needs lots of CapEx, which obviously doesn't fall into OpEx.

IMHO, this is one of the underappreciated aspects of REI and ways to build long term stable cash flows that outpace inflation.

 We all write like Phillip II of Macedonia and while Mike replies like Lycurgus

 Would you mind sharing some of your rebuttals to seller questions? Much appreciated