Opinion: Why NZ house prices are poised to plummet 25% 'like Wile E Coyote'

By Philip O’Connor Will NZ Housing follow US Housing Down? Can Monetary Policy/Inflation Save the Housing Market? In a December 2008 blog post I discussed the apparent relationship between NZ house prices and US house prices, based on statements made in both the NZ and US press. Public discussion of house prices and press accounts with statements such as “houses can only go up in value” were remarkably consistent, although I guessed that NZ was about 1½ years behind the US. Statistical testing, in fact, suggested a 16 month lag, and the closeness of the relationship is remarkable. The size and rate of the increase in house prices in both NZ and the US have been amazingly similar: both experienced a doubling of house prices in 6 ½ years. However, in my blog post, I forecast that if the relationship continued, NZ house prices would fall by approximately 35% by the end of 2011. The initial impetus in the US housing market was fed by a number of sources, including legislation, easy money, securitization, lax regulation, and the banks' ability to sell derivatives and shift risk. The banks’ ability in the US to sell risk to third parties was facilitated by ingenious inventions of investment bankers such as MBS (Mortgage Backed Securities) and CMOs. Banks could also bypass regulatory requirements with their ability to use off-balance sheet vehicles. A transmission mechanism between the large US real estate market and the minuscule NZ real estate market is not hard to imagine. Rising prices in the United States would cause travellers and shrewd investors to surmise that real estate would also rise in other Pacific Rim locations; if a beachfront property sells for $800,000 in California, a similar property for US$200,000 in New Zealand may be worth a lot more. After the initial speculation gains momentum (and there were many stories of foreign investors snapping up beachfront property in NZ over the years), then local residents might catch on and tag along, creating the 16 month lag in the data. In other words, we have a worldwide real estate market. Coincidentally, the NZ housing market also experienced some of the same aberrations in bank behaviour. The traditional 20% deposit was relaxed so that smaller deposits, or even 0% down, allowed one to buy a house. While US bank lending standards were vanishing, NZ banks and the finance sector learned about these new behaviours from the US, and followed by lowering their own standards. Housing bubbles have appeared in many other places in this time including Australia, the UK, Ireland, Spain, China, etc. Robert Shiller has stated that almost the entire world has experienced a bubble in the prices of housing, and he concludes that “Higher home prices in a given country may tend to create conditions for falling home prices there in the future.” (See for example). Therefore, the dynamics common to a bubble may be influencing both countries housing prices. Note, too, that in both countries the fundamental factors that underpin housing valuation -- such as household income and GDP -- changed only slightly. The fact that the value of houses rose without a corresponding increase in income increases the chances that both countries have experienced a bubble in house prices. Using data from the Reserve Bank website, we can calculate the value of NZ’s housing stock as a percentage of GDP. Up until 2002, NZ private residential homes in total were valued around 2x GDP. A few years later NZ homes were worth 3.5x GDP, and are still worth well over 3x GDP, as shown in Figure 1. This relationship is hard to explain, unless NZ residents are expecting a massive immigration wave or a huge inflation in wages and rents.

Figure 1 shows a time series of the value of aggregate private sector residential dwelling values as a percentage of Annual GDP. 1988 to 2009Q3. Source: Quotable Value Limited, RBNZ estimates The valuation of NZ house prices is also high compared to other countries. Dividing the median house price by the median household income, New Zealand houses sell for 5.7x median income. In the US houses sell for only 2.9 times median income, a level that NZ experienced in the early 1990s. Although Australia has a ratio of 6.8 (the same as Auckland), the UK at 5.1, and Canada and Ireland at 3.7, are all significantly underpriced compared to NZ houses (all data from September 2009). However, a bubble in NZ house prices would explain why housing has grown in value relative to GDP. It is widely accepted that the US had a bubble in house prices since house prices have fallen so quickly (see Figure 2 below), so the similarities between the two countries may be related to “bubble dynamics”, a new frontier of finance research. In a bubble, expectations of home buyers and financial institutions have changed in similar ways in both countries (NZ with a 16-month lag) and house prices have not been driven by fundamental factors that affect valuation. Figure 2 updates the previous blog post where an almost perfect relationship was found between US house prices and NZ house prices. The update reveals two striking features. First, US house prices have followed the the stock price index of US homebuilders almost perfectly. The recent increase in the Case-Shiller index for 7 months in a row is a blip that fits within the boundaries traced by the homebuilders’ index (the homebuilders index is shifted forward 12 months). In fact, the futures for the Case-Shiller index mentioned in the previous post have come very close to predicting the actual values of the Case-Shiller index. This indicates that the futures market predicted that house prices would continue to decline and was able to forecast them with a high degree of accuracy. Housing prices may be predictable.

Figure 2. The left hand axis shows the value of an index of NZ median home value (source REINZ) and the USA Case-Shiller 20 city composite index 16 months before. The right hand axis shows the value of an equally weighted index of the stock prices of the 10 largest market capitalization USA home builders 28 months before. All indexes have a value of 100 at January 2000. 'A striking deviation' The second striking feature is that NZ house prices have deviated from US house prices starting in March 2009. Prices have risen and dipped and are now about equal to the previous 2007 high. A few possibilities explain the deviation since the peak. First, the relationship between US housing prices +16 months and NZ housing prices could have been entirely random. However, this is doubtful. The chances of a worldwide housing bubble driven by entirely random events in each country independently are too small to quote. Second, the median house price data may be reflecting a change in the composition of houses sold. If buyers of lower quality houses are being kept out of the market (say because of tightening bank lending standards), the median will reflect the prices of larger wealthier homes. The Case-Shiller index estimates house price changes by examining the price change of houses since they last sold (same house sales). It does not suffer from a bias caused by the changing composition of house sales. The median house price in the US also rose 101% from January 2000 to its peak in March 2007, some 9 months after the Case-Schiller Index peaked. From their respective peaks until December 2009, the Case-Shiller index is down 29% while the median house price has declined only 12%. So although the US median price index rose in step with the Case Shiller price index on the upside, it took longer to turn down and then declined at a much slower rate. This indicates that the REINZ’s median house prices may be underestimating the weakness in the NZ housing market. Stratified median house price data available from www.interest.co.nz also indicates that REINZ median house price is understating the decline in NZ house prices since the peak. In fact, the current median US house price at $NZ363,571 is almost the same as the median NZ house price (using the current exchange rate of $US0.70/$NZ). As income and GDP in the US is much higher, NZ house prices again appear to be overvalued compared to US house prices. 'The China effect' The third possible explanation is that the NZ housing bubble is being driven not just by the US housing bubble but also by the bubbles in Australia and/or China/HK as well, which also may have been caused by US government and monetary policy and which have also not yet burst. While writing this post I came across a blog analysing the Australian housing market. Australia appears to be looking at a similar decline in the value of their houses. For further information on China’s housing market, try a google search on “Chinese housing bubble”. Fourth, the explanation for the recent slight increases in both house price indexes may be explained by the change in borrowing rates, as both the US and NZ housing markets have experienced lower mortgage rates (see Figure 3 for NZ mortgage rates). In the US, mortgage rates have remained fairly low in recent years and the reduction in rates has lead to a feeble bounce. In spite of the US Fed buying trillions of dollars of mortgage backed securities and bailing out owners of mortgages, house prices have only risen slightly. However, in NZ, floating mortgage rates decreased sensationally from 11% to 6%. The huge decline in rates has kept house prices basically unchanged. Another previous blog post challenged the idea that lower mortgage rates would lead to higher house prices. This was based on the observation that mortgage rates had been increasing virtually in accordance with the rise in housing prices. I challenged the causality of lower rates leading to higher prices, if mortgage rates are a casual factor and lower mortgage rates increase house prices, shouldn’t higher mortgage rates decrease house prices? Using the same rationale, mortgage rates cannot be a causal factor if house prices rise regardless of whether mortgage rates rise or fall. I argued that mortgage rates are a consequence of demand. The subsequent behaviour of house prices and rates appears to be a classic example of “buy the rumour, sell the news” so often seen in financial markets.

Figure 3. The Relationship between NZ house prices (right hand axis) and variable mortgage rates (left hand axis). The House Price Index is the median price of NZ houses as published by REINZ. The mortgage rate is the Variable First Mortgage Housing Rate from the Reserve Bank of NZ website. So far, the decrease in mortgage rates has had little effect on NZ house prices, as Figure 3 shows. The dramatic decline in mortgage rates is associated with a fairly small increase in prices compared to the period from 2000-2007. Maybe it helped? However, the fall in mortgage rates may have had some benefits to the housing market. Comparing NZ house prices to US housing prices, the lower mortgage rates may have prevented NZ house prices from following the same path as US house prices, which fell by 25%. So what lower mortgage rates may have achieved is a delay in deflating this bubble. The US housing market declined only after the volume of sales declined significantly. If we begin to see a serious drop off in the volumes of houses sold in NZ (which we may have already) then economic reality may again start to influence NZ housing prices. The final and most plausible reason why NZ house prices have not followed the Case-Shiller index, is that NZ housing prices are about to catch up, with a fall in price of 25% or more. In Figure 2, NZ house prices do look reminiscent of Wylie E. Coyote after he runs off the edge of a cliff and pauses in mid-air just before he realizes he is about to plummet into the canyon. Of course, this need not happen. A stagnation of the housing market for many years combined with inflation could also return NZ house prices to traditional valuation levels. I also decided to look for information in the NZ stock market that may be predictive of future housing prices, as the homebuilders index (ticker: XHB) has been predictive of US house prices. Although NZ does not have a publicly listed building sector as in the US, it does have Fletcher Building Ltd. If the NZ housing market is poised to rebound, we would expect to see strength in Fletcher’s stock price. In fact, there seems to be a relationship between Fletcher’s stock price and NZ house prices from 1998 to 2007, as shown in Figure 4.

Figure 4. The Stock Price of Fletcher Building FBU (right hand axis) and the median price of NZ houses (left hand axis), Mar 1996-Feb 2010. Source: Datastream and REINZ.

Since 2007, Fletcher has departed from NZ house prices, falling 65% from peak to the March 2009 low. It has bounced along with housing since then, but remains at levels significantly beneath its high. Although Fletcher has many business segments and the relationship between house prices and a single company’s stock price (rather than an index) is likely to have greater error, Fletcher’s weakness in the stock market does not bode well for increasing NZ house prices, as the stock market does not see boom times ahead for Fletcher Building. This is the same relationship that existed between stock market prices and house prices in the US before US house prices began to decline. There has also been academic research on NZ house prices using rigorous statistical techniques. Fraser, Hoseli and McAlevey (2006) produced a research paper for the Swiss Finance Institute and concluded that NZ house prices were 25% overvalued at the end of 2005 as warranted by real disposable income. As house prices have risen much faster than disposable income since then, the overvaluation would be more pronounced now. The last time that actual house prices equalled fundamental value was at the end of 2002. A copy of that paper can be downloaded here. Can Monetary Policy/Inflation Save the Housing Market? Although it appears that NZ houses are overvalued, the most important issue now is how the overvalued housing sector will play out in NZ. Every homeowner understandably becomes a cheerleader for higher prices. In talking to home buyers over the past few years, I sensed they were motivated not by greed, but by the fear of missing-out on the window of opportunity to purchase an affordable house if they did not act immediately. Fueled by the belief that house prices would never decline, these expectations and behaviours have helped to create a bubble in prices, and to date have been self-fulfilling. History has trained NZ investors to believe that real estate is an optimal way to protect one’s savings. The dilution of the value of the fiat New Zealand currency has been substantial. Take one very simple example: a 1946 NZ florin, which is the equivalent of $0.20 today, sells for around $4.50 due to its silver content. This means $0.20 now buys 4.44% of what it did in 1946, reflecting an inflation rate (decline in purchasing power) of 5% per year. Inflation has bailed out real estate investors in the past. The overvaluation of housing is resolved by an increase in prices, wages and rents that restores house price/income ratios. Although the real value (adjusted for purchasing power) of homes and equity invested in houses fall, the real value of mortgages fall as well. As long as inflation does not cause havoc in the economy and interest rates can be kept artificially low, homeowners can continue to make their mortgage payments and ride out the decline in real value. (Banks would be the losers in these circumstances.) Real estate investors who could service their mortgages in the 1970s and 1980s (floating rates peaked at around 20%) found that the real amount that they owned on their mortgage was marked down by inflation. The Inflation effect Debates about whether inflation or deflation will occur are a flooding financial blogs at the moment. On one side, writers talk of the unprecedented actions of Federal Reserve Chairman Ben Bernanke, who has doubled the amount of US dollars in existence (the monetary base) since 2008. These writers fear that Bernanke will cause a Zimbabwe-like hyperinflation where the US dollar becomes worthless. Bernanke, who extensively studied the deflation of the 1930s, has the philosophy of simply giving away free money or loaning it out at negative real interest rates. His extraordinary statements include dropping new money out of helicopters to stave off deflation; hence the nickname “Helicopter Ben”. However, his helicopters stay close to his banking friends on Wall Street. But to use an example close to home, the concept is exactly the same as a NZ TAB "bailout day", where all losing tickets are refunded for 100% of their cost. Indeed, this is what Bernanke has done for MBS investors (and Cullen did for NZ finance company depositors). Again, just as the NZ housing market followed the US, the government and RBNZ are strongly influenced by the US policy actions. One only need look at the interest rate policy of the RBNZ following the Fed’s interest rate cuts in 2008 It is quite interesting to note that in the 9 months from September 2005 until May 2006, silver rose 146% and gold rose 84% in NZ dollars. Bernanke was officially appointed to take over from Alan Greenspan in October 2005, so the rush into gold and silver and out of NZ dollars coincides with his appointment. As it was widely expected that Bernanke would attempt untried (and radical) measures to prevent deflation, this does not seem to be a coincidence. Gold in US dollar terms also rose powerfully, surpassing significant past highs. This is the type of behaviour one would expect if gold buyers were scared of Bernanke creating hyperinflation, as gold is widely thought to be the ultimate inflation hedge. Although, like the relationship between house prices and mortgage rates, the gold price does not react logically or deterministically to inflation (gold declined in value throughout the inflation of the 1980s and 1990s) over the long term gold appears to retain its purchasing power. (For example, Alan Greenspans argues gold is an inflation hedge and more here) Gold in them thar houses? It is quite revealing to compare NZ house prices to the price of gold. Figure 5 shows the number of troy ounces of gold required to buy the median valued house in NZ. In 1992 it took 150 oz. of gold to buy a typical house. House prices then steadily increased so that in 1995 a typical house cost over 450 oz. of gold. The increase in the gold price surrounding Bernanke’s appointment has lowered the price of NZ houses in terms of gold, with around 250 oz. now required to buy a typical house. In terms of inflation hedging, NZ houses appear to be very reasonably valued. Of course, Figure 5 begs the question of whether gold is fairly valued. Perhaps the gold “bugs” who fear hyperinflation have bid up the price too high. However, if NZ home owners have bid up the price of houses based on inflation fears, then the high price of houses in fiat terms may be a rational response. Figure 5. The number of troy ounces of gold required to buy the median valued house in New Zealand. January 1992-January 2010. Source: Datastream and REINZ. On the other hand, other financial writers believe that deflation, or a fall in asset values and a rush into fiat cash will occur, and that the debt in the economy is so large that debt levels are unsustainable. Unless there is some way to renegotiate the debt that does not cause further disruption, once a decline in economic activity or defaults on debt occur (as experienced in 2008), the pressure to pay back debt becomes so great that people will sell assets to raise cash and drive down the price of assets such as gold, stocks, and real estate. This will cause further defaults and further deflation and so on. The deflationary writers believe that deflation cannot be avoided, and if it is delayed by accommodative monetary and fiscal policy the final resolution will be worse. This belief is summed up in the following quote: "The wavelike movement affecting the economic system, the recurrence of periods of boom which are followed by periods of depression, is the unavoidable outcome of the attempts, repeated again and again, to lower the gross market rate of interest by means of credit expansion. There is no means of avoiding the final collapse of a boom brought about by credit expansion. The alternative is only whether the crisis should come sooner as the result of a voluntary abandonment of further credit expansion, or later as a final and total catastrophe of the currency system involved." von Mises Human Action (Chapter XX, section eight) Deflation risk NZ households appear to be positioned for inflation and are extremely vulnerable to recession and deflation, as a decline in the value of houses would lead to houses worth less than the mortgage, at least for recent buyers. New Zealand household debt levels are among the highest in the world. Unfortunately the first signs of deflation are starting to appear in the monetary aggregates. The 1-year decline in M3 (money plus credit) of 4.5% in January 2010 is probably the largest decline since the Great Depression of the 1930s. There is hardly a yearly decline in the available data that goes back to 1961 (although data for 1987 could not be found on the RBNZ website), so any decline in M3 is quite significant. Will the RBNZ try the same policies as the US Fed? Given that guarantees on finance company deposits and interest rate cuts followed similar US actions in 2008, it appears likely. Although the RBNZ is being disciplined compared to its peers around the world, it still controls a fiat currency, and it may be that the RBNZ has less power than it thinks. The RBNZ holds substantial US dollar assets and little gold as reserves (gold appears to be only held indirectly through special drawing rights of the IMF, and the IMF is owned by the titans of the global financial system). The policy issues facing Western governments have never been trickier. Whatever they do, they will create winners and losers. The government cannot create wealth, but only transfer it from some to others. For example, with the level of debt so high, bailouts have the potential to transfer gigantic amounts from groups such as savers and future generations to current home owners and banks. Or will only banks benefit? And even with these extraordinary transfers of wealth, it is unclear that any government policy can be successful. Remember none of the experts in government could foresee the 2008 financial crises. How can we expect them to foresee what their actions now will cause in the next few years? In Summary The NZ financial system is strongly influenced by practices of the US (and worldwide) financial system. The overvaluation of the NZ housing sector relative to income appears to be a rational response to (hyper)inflation fears assisted by bubble dynamics. Because their debt levels are so high, New Zealand homeowners are extremely vulnerable to deflation that would be caused by the default/repayment of existing debt. Extraordinary and unprecedented action by governments around the world as they fight deflationary forces of too much debt, by adding more debt, has created uncertainty over who will be the winners and losers over the next few years. At risk is the fiat currency system that is currently being used. * Philip O’Connor is a Senior Lecturer in finance at the University of Auckland’s Department of Accounting and Finance.