I don’t know about Wikipedia,which I consider to be a source of no value at worst and little value at best, but CAPM and the Efficient Market Hypothesis are polar opposites. The EMH states, in its strongest form, that no matter how much information you have that you cannot predict a future return–it’s already priced into the market. CAPM is a system that uses history to attempt to predict future returns. (No, I didn’t go read what’s on Wikipedia–there is no use to read bad information)

I’m saying you can be picky but still have little to no competition in a seller’s market. It depends on whether or not your type of pickiness is popular.

By Kary L. Krismer @ 7:

RE: Indy @ 6 – I’d say it depends on how picky you are about houses. If you only like less than 5% of the houses of a given type, it can be very difficult to buy in even a normal market, because the best houses will go fast. If you’re not picky at all, you can buy in any market.

Great point. It’s possible to be picky and still be able to buy in any market. I like less than 5% of houses of a given type, but there’s not a lot of competition for them. E.g., seems not many folk prefer an old run-down kitchen (I do, since I want to replace it, and not with granite / stainless).

Oh, clearly it’s possible to buy in any market, but it’s nice to be able to look at the house on your schedule rather than having to look at a house within 24 hours of it going on the market and then make on offer on it with the concern being there will be another offer coming in.

As long as the market price is lower than what your personal value in the property, then you will buy, no matter if the market is at bottom or top.

Not saying much, because for many if not most people the personal value is based in large part on a prediction of future market price. For example, if I think a property will be worth 20% less than today’s price at the bottom, my personal value of the property is about 20% less than today’s price (the sooner the bottom the closer to 20%). That sentiment largely fueled the bubble: people valued houses higher because they thought next year’s prices would be higher still.

RE: Indy @ 6 – I’d say it depends on how picky you are about houses. If you only like less than 5% of the houses of a given type, it can be very difficult to buy in even a normal market, because the best houses will go fast. If you’re not picky at all, you can buy in any market.

Great point. It’s possible to be picky and still be able to buy in any market. I like less than 5% of houses of a given type, but there’s not a lot of competition for them. E.g., seems not many folk prefer an old run-down kitchen (I do, since I want to replace it, and not with granite / stainless).

One of the assumption made in the derivation of CAPM is the Efficient Market Hypothesis. (see #7 in http://en.wikipedia.org/wiki/Capital_asset_pricing_model). I don’t see how CAPM is a method of trend analysis. It takes the probabilistic descriptions of the investments as a given and develops a recommended portfolio to minimize risk for a given return. In fact, strictly speaking, CAPM only refers to two points in time: the present time and some future point in time at which the returns occur.

“In terms of a single house purchase, however, I don’t see how else it could be used, as we don’t have a portfolio or a history to develop a beta.”

What you need, according to the assumptions of CAPM, is the variance of the future price and the covariance with other investments. That is far removed from what most people consider when buying a house, although it really shouldn’t be. The portfolio for most people will contain either 0 or 1 house, which also makes things difficult.

I have no problem with attempting using a generalized definition of alpha. The reason I replied is because that your usage falls short because it le
d to your statement

“As long as the market price is lower than what your personal value in the property, then you will buy, no matter if the market is at bottom or top.”

If you expect the market is falling, then it is unlikely to be a good time to buy. It’s called deflationary expectation. Your expected return on a house at that time is likely to be negative. The only way that would have a positive alpha in the CAPM sense in a falling market is if the house has a strongly negative beta. I doubt negative betas occur in the housing market. Perhaps for houses in close proximity to gold mines.

However, if the house has a positive alpha as you have defined it, a person would be tempted to buy. But if that person doesn’t think the market is at a bottom, then they have to trade off how much they expect the market to fall versus the extra hassle of moving into a rental and living there until the bottom.

The notion of a falling market doesn’t usually come up where CAPM is generally used (apart from backward looking commentary). That is because the stock market is highly liquid. Trends do occur in the housing market because transactions take many months, the transaction costs are very high, and you can’t short sell a house in the stock market sense of the term.

Really funny.

Very funny.

I forget who said that Real Estate is a meat head’s investment, which it is. I think you have to be extremely stupid to invest in real estate. Buying Real Estate has always been and always will be a losing proposition.

It’s work no matter how well you think you are buying.

In my opinion that’s what happened in the market place. People bought property expecting a return on debt. You can leverage debt on a better bargain, but that has traditionally taken some skill. Since 1998 Real Estate has been a victim of irrational appreciation. Those in the business never minded until today.

Today we are back to Real Estate as paying off the debt and creating value by manual labor. Once that sinks in people will be getting rid of the Real Estate portfolios faster than a cat with it’s tail on fire.

There are no more future dollars, because we have too many bills to pay.

There are some great accounting concepts in your post. Boeing has some property that they have not realized any increase in value on, as they are not allowed to “mark to market.”

Historic costing.
Mark to market.
Principal balances.
Depreciation.
Change in market value.
Appraisal versus true market value.
Book Value

All of these have a good deal of complexity, yet each one seems so simple.

My favorite term: “Instant Equity’
(Compare “Instant Equity” to historic costing…)

Are we back to a random walk?

“The alpha in CAPM is measured as an annualized rate of return, but the difference between value and price would be measured in dollars, for example.”

Not always true. This depends on the metric space we are using. If you are using rates, then the difference would be the excess positive rate (this alone is difficult to discuss, as we have all kinds of metric issues here too.)

For example, I want 10% on my money, but I have a risk-free opportunity that is 11%. Clearly there is some alpha here, but the amount is another question. We could measure it in dollars, if we knew the base (we still need time in here too, and compounding periods become another issue). We could measure it in terms of excess return, either % or difference in rates (i.e. 10% excess return, or 1% more than 10%). In CAPM you might have an alpha of (depending on how the model is set up) 1.12, or 12% excess return. (Once again it almost goes without saying that we have metric space issues here).

Now that being said, when I do a quick ixquick search, I see that the CAPM definition is quite prevalent. Therefore I think I should have been much more careful and defined how I use alpha. In terms of a single house purchase, however, I don’t see how else it could be used, as we don’t have a portfolio or a history to develop a beta. Of course CAPM is a trend analysis, which runs counter to the Efficient Market Hypothesis (use your favorite form).

I generally use alpha to mean “excess return,” and then the technical computation will follow. (I think this will suite all uses of the term, at least generically, including the various CAPM definitions. The specific computations are yet another issue. There might be a technicality that my generalization does not capture.)

Now who was it that first found the limit of compound interest to be related to e? (the limit is e at 100%)

RE: jon @ 31 –

“when they know the price is going down”

This is the irrational assumption made. If they don’t know the price is going down, and they don’t consider the future value, can it be a rational purchase?

It’s actually just a false premise. There’s no way they can know the price is going down. They can know it has gone down in the recent past, but there’s no way to know the future.

RE: Kary L. Krismer @ 24 –

Are mortgage payments as real as rent payments?

Yes, but the post I was responding to was adding in a drop in value to the cost. It’s not like an increase in value through 2007 did someone a lot of good who bought in 2005 and still owns today. That wasn’t money in the bank. And it’s not money lost for someone who bought in 2007 if they don’t need to sell today.

RE: jon @ 31 –

Alpha, in the most basic sense, is the difference between the value a person places on an asset and the purchase price. This term is then given various forms in different models, such as CAPM. Essentially we are talking about the same thing, as you would not buy an asset, or group of assets, without a positive alpha in the CAPM model. The computation is different.

If I value a bottle of beer at $2 and the cost is $1.75, then I have realized 25 cents in alpha, if I make the purchase.

I’ve never heard alpha used that way. The alpha in CAPM is measured as an annualized rate of return, but the difference between value and price would be measured in dollars, for example.

My recollection, confirmed by http://www.moneychimp.com/articles/risk/regression.htm, is that alpha originated from its usage in statistics.

People throwing around financial terminology without really understanding what those words mean contributed to the mess we are in.

“when they know the price is going down”

This is the irrational assumption made. If they don’t know the price is going down, and they don’t consider the future value, can it be a rational purchase?

Alpha, in the most basic sense, is the difference between the value a person places on an asset and the purchase price. This term is then given various forms in different models, such as CAPM. Essentially we are talking about the same thing, as you would not buy an asset, or group of assets, without a positive alpha in the CAPM model. The computation is different.

If I value a bottle of beer at $2 and the cost is $1.75, then I have realized 25 cents in alpha, if I make the purchase.

Alpha does not depend on a random walk assumption. For example, let’s assume you have lots of money. Now place a value on a property. You will buy if the value is higher than the asking price, as there is a positive alpha. It does not matter how you valued the property–simple random walk, Brownian movement, white noise, or whatever statistical methodology that has or has not been used.

I have a friend who values “community” highly, and she considers ownership an investment in the local community. If asked, “Would you buy in a down market,” the answer probably has nothing to do with finance. She can have positive alpha even in a down market, as her valuation methodology is not based on future market values. Of course in economics we always assume rational players, and this type of play is, simply put, not rational. That said, we know her alpha is positive, as the sale was made.

Also every buyer in the market today has a positive alpha on the purchase.

Also if you go back in the threads, you will note that I did a cash for clunkers deal. My valuation was in part based on the current market value, but I also valued the present value of car in the long run, discounted out, of course. What is reliable transportation worth? (Under factory warranty)

Are mortgage payments as real as rent payments?

RE: Kary L. Krismer @ 24 –

Buyers seek alpha! As long as the market price is lower than what your personal value in the property, then you will buy, no matter if the market is at bottom or top.

If the rate of increase is greater than your discount rate, or cost of capital, then you will buy, even if the market is not at the bottom.

Alpha only makes sense in a market where prices follow a random walk. Real Estate prices follow clear trends for many months. During a down trend, people should hold off buying as long as they can, even if they think it is a long run benefit to them to buy already.

The question now is whether the monetization that the Fed has done this year is enough to reverse the positive feedback in the market to make it start going up, or whether the downward cycle of defaults and resulting falling bank reserves will continue. The 50% rally in the stock market suggests many people think that the cycle will reverse, where rising equity in some markets will reduce the rate of default enough to enable banks to feed money back into the economy, which will then cause prices to rise in the remaining markets. This time around the cycle, the Fed can slow the positive feedback by calling back the reserves it has exchanged for toxic loans. But with Obama’s planned $9 trillion deficit, there will be pressure to keep the economy pumped up for a long time to pay those bills, thus triggering another collapse later on.

Buyers seek alpha! As long as the market price is lower than what your personal value in the property, then you will buy, no matter if the market is at bottom or top.

If the rate of increase is greater than your discount rate, or cost of capital, then you will buy, even if the market is not at the bottom.

Very true and that is the idea of keeping interest rates low. The mortgage payment becomes seductive. At the same time you are correct that rents are artificially high. Mortgage payments have pushed up rents and there was a lack of rental housing. That will go on for a while.

I’ve been caling this the year of the Obama Hope.People are hoping the worst is over, and it is. The free wheeling days of paper trading are over. That’s the good news. The bad news is that all that paper represented debt that now needs to be paid by some entity. Those entities are hoping it will be you.

Let’s make this simple. The best deal you can make is with a seller who has a lot of equity. Maybe they owe $120K on a $400K home. That is called untapped equity. Once the house is sold the money changes hands and has a chance to enter the economy.

The seller, banks, lenders, tax people, and neighbors are all very excited to have the home sell for as much money as possible. The only person who wants to pay the least is you, the buyer.

So there is a collective holding of breath hoping that pricing will remain high.

Find a place to stay for another couple of years. After 2010 the housing market will be a distant memory and people will just be buying houses again.

RE: Urban Artist @ 18 –

It’s not a question of how much the rent cost, but rather how much further is the value going down?

Let’s say $2,000 rent gets you a place with a market value of $500,000. If the $500,000 place goes down 10%, then you will have ‘paid’ $50,000 in lost value. Of course there needs to be present and future value computations.

The difference is, rents are real, future values are not, unless you sell (or want to sell) at that point in the future.

A decision to buy at the bottom is a decision not to ever buy.

Home value population:

{750k, 675k, 607k, 546k, 492k, 442k, 398k, 358k, 322k, 290k, 261k} (each one 10% less than the next higher)

Round 1:

These homes sell at full value:

{675k, 607k, 546k, 492k, 442k, 398k, 358k, 322k, 290k, 261k}

Round 2:

These homes sell at 10% off previous price:

{$750k, 675k, 607k, 546k, 492k, 442k, 398k, 358k, 322k, 290k}

Note that the selling prices are exactly the same as round 1.

Paired data: Down 10%
Non-paired data: no change.

The sales mix did not change.

Now with some minor changes in the sales mix, we could obtain all kinds of different results. Of course we could have major changes in the sales mix too.

Also note that we could play similar games in the other directions too.

I ask: Why does the question of “is this just a seasonal (etc etc)” keep coming up here when seasonally adjusted data either exists or can easily be generated?

Without doing much thinking, I bet we could play similar games increasing housing prices (paired data), yet the average stays the same (not paired), and the tiers either increase or decrease.

It’s not a question of how much the rent cost, but rather how much further is the value going down?

Let’s say $2,000 rent gets you a place with a market value of $500,000. If the $500,000 place goes down 10%, then you will have ‘paid’ $50,000 in lost value. Of course there needs to be present and future value computations.

In any event, don’t get caught in the trap of simply comparing monthly rent to “mortgage payments.” The future market value is key these days.

(By the way, my former landlord purchased at $285k. Now places are selling for under $200k. He offered to sell it to me for over $300k. According to the county records, and making some nice assumptions, he owes about $300k on it. It was a 100% financing purchase, and then he refinanced as “values went up.”)

In any event, if you think market values are going to go up rapidly, buy today!

I can’t believe it.

People are paying way too much for properties. I saw a good deal about four months ago in Magnolia. The rest of it is complete insanity. That’s where you do what you’ve always done.

The sales data is deeply flawed. Just because people pay too much doesn’t mean the value is there.

As far as appraisals and mortgages are concerned, anything they turn out today is much better than they have from the past three years.

RE: The Tim @ 12 –

What happens, and I do not know the answer to this question, if all homes decrease in value by 10% (paired methodology), yet the average purchase price (not paired) stays the same?

Good question.

Lets say in period one you sold 90 houses in a “normal market”, 10 each at the following prices: $100K, $150K, $200K, 250K, 300K, 350K, 400K, 450K, and 500K.

The “tier breaks” would be at $225K and 375K, and the “average price” would be 300K.

Now, lets say in the next period all prices drop 10%. They are now $90K, 135K, 180K, 225K, 270K, 315K, 360K, 405K, and 450K, but that the “sales mix” differed so the average remained 300K. There’s a lot of ways you could come out with that result, but one would be to sell 7 homes each at the 90-225 levels, 10 homes at 270, and 13 each from 315 to 450.

The tier breaks would now be somewhere a little higher than 225 and a little lower than 405.

So, the “average price” stayed the same, and the “tier breaks” rose, all while all prices fell 10% That’s the power of the sales mix.

I think I need to go read the CS methodology.

I am guessing what you’re saying is that the average prices are essentially staying the same while the 33rd and 66th percentile of the population is increasing.

What happens, and I do not know the answer to this question, if all homes decrease in value by 10% (paired methodology), yet the average purchase price (not paired) stays the same?

Percentiles are by 1% increments (0, 1, 2, 3, …, 50th (median), …, 99th, 100th) (First percentile is normally 1%, where 0 is called “minimum”, and similarly the 100th is normally called the “maximum”)
Quartiles – 0, 25th percentile, 50th percentile (median), 75th percentile, 100.
Deciles – by 10 percentiles (0, 10th, 20th, …, 90th, 100th)

The maximum is the 100th percentile and the minimum is the 0 percentile. The range is the maximum less the minimum.

I don’t normally use 3rds, and I am not sure if a special term exists. (33rd percentile, 66 percentile)

That’s not the point, however. The issue is that if the observation points (“breakpoints”) increase, then why does that imply a mix in the sales?

Could you clarify this implication? How do you know that prices did not change rather than the change in mix, or some combination of the two?

Sorry, I should have added some more explanation of this statement. None of the tiered indices increased more than 0.8% from last month, but the tier breakpoints bumped up 1.2% and 1.3%. An increase in the index for a given tier indicates rising prices. If the increase in the tier breakpoints is greater than the increase in the index, that implies that the sales mix is shifting toward the more expensive homes.

The Tim-

“The tier breakpoints continued to shift slightly upward in June, implying a shift in the sales mix of homes away from the low end toward the high end.”

Could you clarify this implication? How do you know that prices did not change rather than the change in mix, or some combination of the two?

A better way to measure the change in mix would be a graph showing CSI and the 3 month moving average of NWMLS prices for the same geographic area. Even that would not be exact, because CSI exludes new homes and large price changes, and over-weights recent sales, which in this market are distress sales.

I’m not sure what the term is for such cutoffs. But it’s like having two medians, one and 1/3 and the other at 2/3.

On the other hand, I’m not sure how useful the information is, because I’m not sure whether it’s 1/3rd of all sales, or just 1/3rd of their accepted pairings.

I’m enjoying my recent rent cut and watching the madness from the sidelines.

“The tier breakpoints continued to shift slightly upward in June, implying a shift in the sales mix of homes away from the low end toward the high end.”

Could you clarify this implication? How do you know that prices did not change rather than the change in mix, or some combination of the two?

Also do you have a list of the historic break points? I am guessing that the break points went up in the now established bubble period.

At the same time I would suggest that if you are a potential home buyer, there’s no reason to rush out and buy right now if you don’t have to. There are a few good deals out there right now, but they should continue to be out there for the next couple of years.

This is a key insight. When folks with stable jobs and decent savings who were priced-out of the local market and have pent-up “American Dream” home-ownership demand finally see a great house drop into their price-range, the temptation to buy now is almost irresistable, especially if you think the $8K credit won’t be extended.

But there are still risks. By my estimate, similar deals wont disappear any time soon, so you have little to lose except maybe another year in a rental, and you may gain up to another 20% discount if prices continue to fall. On the other hand, if you buy now, put that 20% down, prices continue to decline for the next year, and then some life-event occurs and you’re forced to sell, then you could lose your entire down payment.

A mortgage is a leveraged investment, so it’s better to “miss” the bottom and pay 10% more than to anticipate the bottom wrongly by 10% and lose half your down-payment if you suddenly have to move. Rent, wait, and see seems to me to be the sensible strategy.

Time for society to finally work 2 jobs again!

http://www.youtube.com/watch?v=ydVp-2B8qlw&feature=PlayList&p=E0E0372E75DC42D2&index=50

Not.

The most cursory glance at YOY foreclosures, bankruptcy and unemployment stats refute the propaganda.