Home : Blogs : Tim Maitski "Video Agent Guy" : Tim's Atlanta Real Estate Blog ~ Save Money. Avoid Rip Offs.

When Is a 20% Drop in Average Prices Really a 0% Drop in Average Prices?

Market stats can be deceiving.

Every month the MLS in Atlanta comes out with average home sales prices for the Atlanta metro area. They are useful in general terms but let me show you how easy it is to get an entirely incorrect picture of what's going on in the Atlanta market.

Fulton county is in the middle of metro Atlanta and is a very long county. It could take you one hour to drive from the south side to the north side. Houses in the north part of the county might cost on average $500,000 while houses in the south part of the county might cost on average $200,000.

Here's a hypothetical situation with simple numbers to make the math easier.

10 sales at $500,000
10 sales at $200,000
= ave. price of $350,000

Next year the prices stay the same, total number of transactions stay the same, but the relative number of transactions changes in the two areas.

5 sales at $500,000
15 sales at $200,000
= ave. price of $275,000

The MLS would report that the average price of a home in Fulton county dropped 21.4%, from $350,000 to $275,000 and total sales remained constant at 20.

That would be an amazing number that would be reported in the newspaper. If you are an agent you might recommend that a seller reduce their price, when in fact the only thing that has happened is that more homes sold in south Fulton than north Fulton.

I've spent many hours this week creating market reports that drill down to smaller market areas around Atlanta. I've created average house price charts for specific areas going back to 2002 which show both price and volume. I will start posting them in the next few days.

It has been really eye opening. There are sections around Atlanta that have had average prices drop around 50% in the past year while other areas have remained level. In general, the volume of transactions in the decreasing markets is increasing and the volume in the steady markets is declining significantly. This explains why so many agents are hurting all over. Prices in north Fulton are staying even but there are half as many homes selling. In south Fulton, prices have plummeted and the number of transactions has increased. This will skew the area wide numbers in a similar fashion as my simple example.

The bottom line is real estate is local. Use national numbers with caution and also metro wide numbers with caution. Know the big picture but also know what's happening in your very local area.

Posted by Tim Maitski "Video Agent Guy" on 05/28/2009 10:43 PM

Comments (31)

atlanta market stats

31 Comments on When Is a 20% Drop in Average Prices Really a 0% Drop in Average Prices?

JAN

Nice post Tim. You really bring up a good point regarding statistics as they can often be misleading.

You are so right....and now with bank owned and short sales priced so far UNDER market value, the sales that are showing as higher than the asking price are decieving too. We really have to work with our clients to help them understand this market.

Excellent point to make. I've tried to do the same thing when I've done the local reports. The point I try to make is that the average price of the area doesn't mean that individual homes have decreased in price, it merely means that buyers may be choosing to BUY the lower priced homes. As your chart shows, when that happens, the "average" price goes down.

Way to Go! Tim! Congratulations! Now that is an excellent analysis for the Atlanta public and Real Estate practitioner in Atlanta to utilize. Sure goes to show the adage "Real Estate is LOCAL!" If you have other blogs, I would definitely post this one to them!

But this information could be even more beneficial if you were to let your Sphere of Influence , your farm area know it as well. I think I would even pay for a full page ad(or 2-3 pages) in the Homes magazine for Atlanta, and post it there? That information should be a real eye opener!

Tim,

Really good illustration as to how numbers can be deceiving.

That being said, I only use the median data in order to support values because the average simply is not a good and fair representation.

Tim - it's a good example of how one really needs to look at the local makret, and data are just a part of the story. Unfortunatley that is what gets reported and what consumers here. The same can happen with the median, if enough lower priced homes sell, or higher priced ones do. If does not necessarily prices are really dropping.

Jeff

JAN

Tim- interesting observation. Here in Brevard County which is near Orlando, FL, our foreclosure rate is high: and is having a large effect on the "average sales price". Banks have to sell and are giving away properties mostly to investors like myself, while many homeowners are choosing to pull their homes from the market. In my opinion the current average price (here anyway) does not reflect the true value. Most of our sales are REO's which I consider distress sales, not indicative of the true market.

MAY

Hi Tim -- Your post illustrates why if a sample size is not large enough, and confined enough to direct comparables, macro-level averages are meaningless.

They should always have the "margin of error" posted with statistics of any kind, IMHO.

They're doing something 'new' in our MLS area, showing the daily number of Lock Box entries.

Keep in mind, this is not for my local market area, the Metro Portland region. This is for the entire state or Oregon.

But, it's still interesting, and numbers are deceiving. They don't tell you how many "listings" these correlate with.

I just did a check "look see" on ACTIVE Listings from $0-$9,999,999. There are close to 25,000 current ACTIVE listings in the MLS regions I have on to pull from on RMLS -- a good portion of the ENTIRE State of Oregon. Some areas, Salem for one, has their own multiple listing service, and might not feed into ours. The coastal areas also use their own MLS and might not be a member of our local MLS.

So BIG DEAL! 3,423 is the number of lock box activity for Saturday. And what if, the listing agent went in and used their Supra Key to get the door unlocked. Maybe 10% of that 3,423 were from the agents themselves?

THe bottom line is: There are some GREAT properties out there, and SELLERS are very eager to sell. Go out and MAKE A DEAL!!!

Around here, the paper will print some story making it seem like the sky is falling, then alter the story year end. It is frustrating.

I find that even in smaller cities, we can see discrepancies from one area to the other.

Congratulations on this explanation! I always tell people that any reported "drop" or "increase" in prices in a zipcode does not mean that any single home's value has gone down or up that much... that's it's a reflection of the price point for homes that sell. Your illustration makes that clear in a much easier-to-understand manner.

Hi Tim,

We've seen the same thing here. When reading the median prices most consumers assume thats a price drop across the board, not so.

Nice work on this, Tim! To really get value out of statistics, it's important to understand what factors contributed to a change from one period to another. You've done a great job of finding 'the markets within the market.'

MAY

You really need to know your local market to make since of the data you are looking at. The smaller the area the better the results will be.

Tim, nice way to present it. I have to tell you it is a constant debate at our board. We are in the process of revamping how we do the stat's as in the past we have included all listings listed by Ann Arbor Area Board members, including listings we take outside our county.

With Washtenaw being a high selling county you list a few outside of it and bingo it looks like our average home prices have come way down. It is much better to just look at very hyper local stat's like you have been doing

It is so confusing to sellers and buyers. You want to low ball every home even in a area that is short on supply. And selling close to asking price.

I was wondering why so many comments on an old post. I got featured! Thanks to whoever.

The problem with most real estate statistics is taking the time to figure out where the numbers are coming from and what your potential error size could be. Unfortunately, newspapers and most people just want headline numbers and really don't want to be bothered with the details.

I think that the best scenario might be a more localized Case-Shiller index. They follow individual homes over time but they only cover 20 major cities. It Atlanta, that's just too big and diverse of an area.

But even the Case-Shiller Index can be skewed by a high percentage of foreclosures. If my particular home gets trashed after foreclosure and goes from $200,000 to $100,000, does that mean the values of my neighbors' homes have gone done 50%? It just means that my home has been trashed and is only worth $100,0000 now.

Good reminder to study our numbers carefully. Also if everything is compared to the best years (2005-2007ish) then of course there will be huge differences. Keep a good perspective.

Hi Tim ~ it is much more difficult to track changes in real estate values than most people acknowledge. You need to track very similar properties over time and that's close to impossible.

I thought of Case-Shiller as I was reading your post and then noticed you mentioned them in your comment. I think that they put together their model because of this difficulty of tracking prices. I have a vague memory of reading an article about a study they did years ago. They traced changes in property values in a neighborhood in Holland (if I'm remembering correctly) where they were able to compare like properties over a long period of time. We don't have that luxury typically - properties vary dramatically in terms of level of renovation, condition, site, location, etc etc etc. and we're almost never looking at the same unchanged property over time.

The other complicating factor is that sales figures are really tracking what people are spending - not the type of property they're gettng for the money.

Lots of food for thought in your post.

Liz

Good illustration Tim.

In the end the only number that counts to a seller is what's going on with my house.

NEARBY local data is helpful, national data less so.

This a real good way to put it Tim. I see generalizations for stats that just don't jive for ever situation/area.

Great post.

Just today somebody asked us to for sales stats for a zip code. That zip code covers a lot. I asked what more specifically they are looking for. This is also why we do hyper local reports. ~Rita

The days on market stat drives me nuts too. it is often used to say look at this, the market is getting worse becasue DOM is 15 days higher. it could also be that the homes that extend the period arte overpriced with realtors afraid to ask for price drops. I like to focus on inventory, and real sales.

Many good points. We are constantly reminded that you have to look at the local market.

what a great blog post i talk to our new members all the time and point out the same exact thing. i'm glad you decided to compile your stats on specific neighborhoods instead of alowing your readers and customers to be so easily misdirected. It takes a very savvy agent to notice this concern let alone address it in such a dynamic way. I only wish all of our new members would read your blog it would clearly display the benefits of writing your blogs and market reports on specific neighborhoods instead of for counties or cities as a whole. keep up all the great work tim!

I've been writing about this in LA for 6 months. The number of sale <$750K is up 28% while the number above $1.5 million is off 54%. This has led the media to breathlessly report stomach-churning average price "declines", when much of it is simply the imbalance between a very hot lower end of the market (subsidized rates) and a dead upper end (affluent sellers who can wait it out without reducing their prices). Dana Graham Palos Verdes, CA

Tim... excellent point. You made this statement... "The bottom line is real estate is local. Use national numbers with caution and also metro wide numbers with caution. Know the big picture but also know what's happening in your very local area."

This ticks me off, because we have been seeing this all over. Good example.. just the other day, someone stated that consumer is the highest it's been in 6 years. Okay, where did they get that number from? Are things being construed? Left out? Just to make this sound good, to get the stock market in the right direction, to make rates rise, so we stay away from inflation?

Overall, just so much out there and it's seems to be very easy to make numbers work in your favor. I am speaking to plenty of people who are hurting. Businesses are still closing,... etc, etc.

Okay, sorry, didn't mean to hijack your post. But just that you bring up a good point and I am seeing this to often in the media also.

jeff belonger

Real estate isn't local... it is hyper local. Even ZIP codes are too big in many places. There are so many things to take into account... dollar volume, prices strata, activity at specific price points... It would take a fuzzy computer to deal with it. I guess we call that fuzzy computer an experienced real estate agent.

MAY

This is so true. In the city of Tempe there are areas that may average 100,000 or so and a mile or two away there are million dollar homes. Then, everything in between that range in the same zip code. I agree with the idea of keeping it very local.

A good example of how you can bend the numbers to prove whatever you want them to prove!

Leave a response…

Name:

Email:
(optional)

Comment:

What does the graphic say?