Gold versus Real Estate, a 40yr Perspective

Patrick Bateman

Recently I was interested in investing in gold, and I wanted to determine whether it was a good value or not.

I have crunched the numbers, comparing the value of gold to various commodoties, such as housing, renting, salaries, and the stock market.

This graph may help you decide if investing in gold is a good idea.



This graph compares the value of gold versus the value of real estate, the value of renting, and the value of investing in the Dow Jones Industrial Average. I've also included "M2", which is one of the measurements that the United States Federal Reserve uses to measure the supply of money.

The lines in the graph are calculated by dividing the numerator by the value of gold. For the past fourty years we have used a 'fiat' currency, which basically means that there is not a finite limit on the money supply. There is a finite supply of gold, so we can evaluate investments by comparing them to gold. Which is exactly what we're doing in this graph.

Here's some things that I see in the graph:

• We've lived through a bubble in the stock market that was much longer and larger than I'd realized. The bubble lasted about twenty years, from 1989 to 2009.
• In the half decade from 1996 until 2001, the stock market bubble grew at a faster pace than the money supply. This may indicate that investors were buying stocks with debt, inflating the stock market bubble to an extraordinary degree.
• According to popular sentiment, the real estate bubble peaked in 2006. If compared to gold, the real estate bubble was already deflating for three years. This may indicate that 'the smart money' was already moving out of real estate by 2003.
• When numerated in gold, the cost to rent has fallen precepitously for a solid seven years, from 2002 until 2009.
• In my opinion, the most interesting parts of this graph are areas where the lines diverge. I believe that divergence is an indicator that an investment is good or bad.
• The most notable trend of divergence is the stock market bubble, from 1996 to 2001, where the value of stocks went up nearly a thousand percent in five years(!)
• I've included a metric of stocks, real estate, and salaries. It's the orange line. Based on that metric, we can see that salaries exceeded the average from 1980 until 1996. Nearly two decades of affluence. After that, salaries lagged the average, particularly in the year 2000. But the value of stocks was accelerating at a hectic pace from 1996 until 2000, which may explain why many were compelled to invest.
• Compared to gold, ALL investments have been deflating for a decade. But real estate didn't start deflating until nine years ago, and it held most of it's value until 2006. When compared to gold, real estate was clearly a declining asset in 2004 - it just wasn't declining as fast as everything else was.
• The Federal Reserve's supply of money began to increase dramatically in 1981, and continued to explode until the fall of the Soviet Union. The expansion subsided in the early 90s, but picked up dramatically in 1997. It's interesting to note that the supply of money didn't seem to make a large dent in home prices or incomes in the 80s. In my opinion, this is because it was spent on defense, which isn't an efficient way to increase incomes. In other words, a substantial investment in defense spending during the 80s yielded a modest increase in wages.
• When numerated in gold, everything has been dropping like a rock for a decade. But incomes and the cost to rent have been particularly deflationary.
• As noted on the right hand side of the graph, incomes, rents, and the cost of housing have settled at levels which are comparable to the lows we saw thirty years ago. The money supply is higher, and the stock market is higher. This may indicate that stocks are still overvalued, and that home prices are now undervalued. It's up to you to evaluate the data.

Patrick Bateman

I thought it would be interesting to compare the value of gold against the value of housing, using Case Shiller data. It's interesting to note that it may be possible to predict the peak and the trough by using the price of gold.

For instance, when we look at the Case Shiller data, we see a peak in housing prices in 1979, 1989, and 2006. But the peak in home prices when denominated in gold is much earlier. Note the peaks are in 1975, 1986, and 2001 My hypothesis is that 'smart money' liquidated their real estate prior to the real estate peaks, and then put the assets into something safe, which was gold. Therefore, a peak in asset prices when denominated in gold may predict a peak when denominated in dollars.

Scotsman

A fun exercise, but very dangerous in terms of yielding usable investment data. Any group of variables, divided by an increasing divisor will show a downward trend. But what have you shown in terms of relationships or causality? How many variables have you really captured? As an example, what if you substituted the inverse cost of a gig of computer storage for the price of gold? And what would it really mean?

Going back to your example, how do you know whether it is housing that may be under valued, or gold that is over valued?