Units or Houses - where should you put your money?

In this newsletter we follow up a question a subscriber asked me - are returns better on houses or units? Firstly I'll tell you my prejudices before we did the research. I was confident that houses outperformed units. Why? Because as the old adage goes - as far as land is concerned "they're not making any more"

Thus while you collect your rent, the value of the land your house is on rises as the population grows and land gets scarcer. By contrast, if the price of units rises, they can make more of them. If you check out the skyline of your local capital city, you'll see what I mean.

Nevertheless, this reasoning is not complete. If it were that simple, no-one would invest in units - because they could put their money into houses and make more. If we assume that investors are not stupid and that they're trying to get the best out of their investments, then one would expect that investors would not buy units unless they came with something which compensated them for the lower expected capital returns - like higher rental yields. Typically units do have higher yields.

Still I've always been a 'houses' man so our subscribers' question was a good opportunity to check things out. I'll tell you below how the answers weren't what I thought they'd be. Then I'll tell you why I'm not sure I believe them!



Graph 1 shows the average total returns (ie capital gain plus rent) earned by investors in houses and units in the major cities from 1994 to 2002. Over this period the average unit in Sydney, Melbourne and Perth yielded a higher return than the average house.

If returns were higher for units, this suggests that the risk may have been higher. For one good explanation as to why returns are higher in some parts of the economy - eg. shares - is that risks or earnings volatility is higher. But guess what? It turns out that in our major markets - Sydney and Melbourne, volatility was substantially higher for total returns on housing than on units (Graph 2).



So what does this mean? Since units in Sydney and Melbourne yielded higher returns at lower risk, they seem unequivocally to be a superior investment. In other cities there was a trade-off between risk and return so the better investment would depend on an individual investor's risk preference.

Clearly in our two biggest cities the shortage of land is most acute. And you'd expect houses not units to benefit more from that as land prices rise. So we'd expect house prices to rise faster than unit prices. How can this be reconciled?

An economist is supposed to be someone who finds something that works in fact and tries to work out if it works in theory. It doesn't accord with my idea of how the world works, so I wonder whether there's something wrong with the measurements.

I think there is. Let me explain. I recently moved to Port Melbourne and in the first year I was there Port Melbourne prices were reported as rising by around 70%. Now I can tell you we did pretty well, but our house price didn't rise by 70% that year. More like 15%. What happened was that the price rise quoted in the papers was the median price. The median is a measure a bit like an average - it's the middle point. Half the prices in a series are above the median and half are below.

Now when we moved into Port Melbourne, an old working class suburb, there was a huge development being established there. Before the development commenced, the median price was around $170,000 but the houses in the development were targeting more affluent people than your standard Port Melbourne buyer. Their prices started at around $300,000.

And there were many more houses sold in the development than in the rest of Port Melbourne. So you can see why the median price soared. Here the median was quite misleading, and what was necessary was something we don't have. Proper indices of resale of the same houses through time (with allowance made for depreciation and improvements to the same houses).

How could this mess up our calculations of returns for houses and units? There has been a huge increase in the number of new units in Sydney and Melbourne over the period we are examining, they have catered to changing tastes for living closer to the city and they have catered to an increasingly affluent clientele. So the whole world has been a little like Port Melbourne.

A surge in high rise unit construction is giving affluent young Australians (and 'empty nesters') easy access to the city for work and to the city and inner suburbs for recreation. 'Low rise units' - the kind that are often built further out in the suburbs - have not been part of this picture. We're not building many more of them than we ever did.

So the composition of units coming onto the market has been changing with increasingly well appointed and located units coming onto the market targeting increasingly affluent buyers.

A second explanation for the strong performance of units in Sydney and Melbourne is that our data doesn't go back far enough. Graph 1 shows house and unit prices in Sydney over the last 2 decades. Clearly there is a lot more going on than what we can capture in our 1994 to 2002 window.



Firstly we missed out the effect of the surge and collapse of dwelling prices in the late 1980s. Secondly, in our sample period we capture the surge in unit prices in the periods preceding the Olympic games. This is clearly an exception rather than the rule. However, because our sample period is so short this extraordinary behaviour of unit price can have a disproportionately large effect on returns.

So what should the reader take home from all this? Well the data suggest that between 1994 and 2002 units in Sydney and Melbourne were a better investment than houses. In other cities there was a trade-off between risk and return. However, there are two good reasons to view these results with caution. Its opened my eyes a little. But I suspect I'll keep being a 'houses' man for a little while longer - maybe until I'm an 'empty nester' myself. But I'll keep my eye open for units with scope for appreciation with strong rental yields!

Cheers,


AKA Nicholas Gruen
November 2002

Please note: The observations made here are general and indicative.Nicholas Gruen is not a qualified investment adviser. Further his comments are general and do not take into account your specific circumstances. Nor are they warranted as free from error in any respect whatsoever. You should not rely on any aspect of them without taking independent financial advice relating to your own specific circumstances. We suggest you obtain advice on a fee for service basis rather than from someone who earns either up-front or trailing commissions from investments they recommend. We would be happy to let you know of service providers who provide advice on this basis.