08 May Medians are misleading
We sort out the confusion that surrounds media reports that confuse median price movements with actual market fluctuations.
Kevin: One of the frustrations I’ve had from time to time is how the media constantly plays on medians. If you see a movement in the median, either up or down, they’ll maintain that’s an indication that the market is moving either up or down when it has absolutely nothing to do with it.
Let’s try and clarify that a little bit more for you. I can do that by introducing my next guest, Richard Rossman from Secret Agent and a really interesting article that they’ve put together, which is why you can’t trust averages or even, in this case, medians. I want to talk about the difference between all of those and what that really means.
Richard, welcome to the show, and thanks for your time.
Richard: Thank you for having me.
Kevin: This is long overdue, trying to clarify what medians are. Firstly, in your own words, explain to me why we really can’t rely on medians as an indication about what’s happening to the market.
Richard: Yes, sure thing. Let’s just discuss what an average or a median really is. It’s a metric that tells you if you have all data points, or in this case, all housing sales in your sample and divide that by the total number to get the average or you find the middle number. If you think about it, it’s basically the total money invested in the housing market per house.
Kevin: The way that I’ve described it – tell me if I’m right or wrong here because you’re much more skilled at this than me – the median is really an indication of where the predominance of people are buying. In other words, if more people are starting to buy more expensive properties, then the median is going to move up. It’s not an indication that the market is moving; it’s an indication that buyers’ habits are moving up. Is that a fair way to talk about it?
Richard: Yes, basically. The median, if you think about it, is just the middle number of your data set, so it’s between the minimum and the maximum. Especially looking at the whole aggregate, if you have smaller houses and bigger houses, if they’re moving in different directions, the median won’t pick up on that sort of thing; it just picks up on the very general trend of are people spending more money on houses; it doesn’t tell you about the stock or supply of the actual houses.
Kevin: Let’s take an example because I know in your newsletter, you’ve done that. Let’s have a look at Fitzroy in Melbourne as probably a classic example. Run me through the figures there and that’ll demonstrate the message, I think.
Richard: Absolutely. We looked at Fitzroy housing sales from 2011 until the end of 2015. What we found was that the reported average would be about 8.7%. Those are the big figures that you often see on national TV, on RealEstate.com.au, or other sources like that. But if you compare it to just using houses that were bought and sold multiple times in that time period without any structural changes to the houses, then you’re looking at a growth of about 5.75%, which is a difference of almost 3% in over-statement from the average. One big factor is a building developer coming in, buying a big house or a big block of land and then subdividing it, which adds a lot of value to the market. The average doesn’t segregate that.
Kevin: Then explain to me why we are so focused on medians. What do they tell us?
Richard: I think it’s the simplest and quickest way to get an overview. Often, people don’t look any further because it tells you a general overview of the market right now. It’s is there more demand for houses and how is the housing industry going? Are people spending more money in it or less money in it? But it doesn’t tell you if you have a house, how much it will be worth a year from now.
Kevin: How do you find that out?
Richard: The way you find that out is by constructing an index using just houses that were bought and sold multiple times without any structural changes such as renovations, subdivisions, or anything like that. If you just use those, basically, what you’re getting is the true capital growth and you’re taking out all other factors that are adding value to the housing market, which is reported in averages.
Kevin: Where can the average punter get that sort of information?
Richard: Our website, quite simply. It’s something we’ve been working on recently, especially since it’s not really out there that much. Housing indexes are done on a very big basis – say, all of Australia, for example, would have a housing index. But if you want to invest in a city or even a specific section of suburbs, it doesn’t really tell you enough.
What we’re doing at the moment is we’re working on constructing indexes for suburbs, specifically, and also for cities such as Melbourne that give you much more in-depth insights into the true capital growth.
Kevin: Will that information be available for all of Australia or just for Melbourne?
Richard: Because we’re a rather small team, we’re just going to be working on Melbourne at the moment. But we might be able to do a collaboration in the future on all other areas, as well. It just depends.
Kevin: The website is SecretAgent.com.au. That’ll get you there.
Richard, I want to thank you for giving us your time. We’re unfortunately out of time, but it’s really interesting. As I said, it’s one of my major frustrations, this focus on medians.
Richard: Me, too.
Kevin: The website to go to: SecretAgent.com.au. Richard, thank you for your time.
Richard: No worries. Thank you very much.