What's happening in the property sector?
A speech to the Property Council of New Zealand in Rotorua
The property sector, in the broadest terms, is something that fascinates most people. Most New Zealanders own or aspire to own some kind of property, and property forms a significant part of the wealth of many of New Zealand's households. As a country with a substantial reliance on agriculture, both currently and historically, the buying and selling of rural properties also has a particular potency in our national imagination. The Reserve Bank, by contrast, has a technical interest in the property sector, though of course we too know how perceptions can be everything.
The Reserve Bank is charged with two key tasks, aside from issuing currency. They are implementing monetary policy to achieve price stability, and maintaining the stability of the financial system.
Changes in asset prices, for the most part property prices, can have consequences for both of these tasks. Most asset prices do not directly enter the Consumer Price Index; the prices of new houses and cars are included, but that's about all. However, even changes in the prices of assets that are not included in the CPI can have an indirect effect on consumer prices. For example, a sharp lift in commercial property prices or rentals ultimately impacts on the `cost of doing business'. In turn, that has an impact on the prices faced by consumers.
Changes in asset prices can also have consequences for financial stability. We know from the experience of the late 1980s that a large fall in commercial property prices can result in some large firms running into serious financial difficulties, with major consequences for financiers and banks. It is important that our financial system is robust enough to withstand such shocks.
So, in short, the property sector is definitely on our radar screens. What I want to do, therefore, in this speech is give you an idea of how we view the sector and its various parts.
Of course, one can start with the core economic idea of supply and demand. However, we also know that the property market behaves differently from the market for consumer goods. In the property market, supply tends to be relatively inelastic. Or rather, it tends to respond quite slowly to changes in demand.
In fact, the demand for buildings can increase dramatically as output rises. Consider a single firm. The value of the building stock that a firm uses can be large relative to the annual output that the firm produces. Hence a rise in output can result in a change in demand for building stock that is even larger, in value terms, than the initial rise in output.
We know that the amount of new building work that can be put in place in a short time period is limited. For one thing, new buildings, and even alterations, involve design time, and time to get through the planning process. Second, there is always a limit on the building industry's ability to meet demand. Such bottlenecks have been in clear evidence over the past couple of years as the demand for new housing and apartments has accelerated.
Because supply often lags demand in the property market, there is the potential for a mismatch between the supply and demand, and this can work in both directions. When demand for property cools, due to a slowing economy, it's hard to switch off new supply in the pipeline. So a rapid rise in prices can be followed by quite significant declines later on. All property sectors tend to exhibit price cycles, with these cycles reflecting this mismatch between demand and supply.
Demand for most kinds of properties over recent years seems to have outstripped increases in supply -- prices and rentals across most property classes have generally been moving upwards and vacancy rates downwards. We can attribute much of that strength directly to developments in the broader economy. Whilst it hasn't been all plain sailing, this year the New Zealand economy entered its fifth year of unbroken growth. Just as that expansion has drawn heavily on the economy's surplus labour and productive capacity, so too has it fuelled the demand for property. When you consider some of the causes and consequences of that growth, it's not hard to see why the property sector has fared pretty well over this period. Some of the following statistics may help to put some perspective around the demand for property. Since 1998/1999, when the business cycle caused by the Asian crisis and drought bottomed out:
- the total output of the economy has expanded by about 20 per cent;
- the volume of retail sales has expanded by about 25 per cent;
- export volumes have risen by nearly a third, driven heavily by the primary sector;
- the annual operating surplus in the agricultural sector has risen, in real terms, by around 35 per cent;
- the number of people employed, either full time or part time, has increased by around 220,000;
- net immigration has added 67,000 new people to the normally resident population, and when increases in foreign students and those here on work permits are included, the figure is considerably larger still;
- the annual number of tourists visiting New Zealand has risen by nearly 1 million; and
- there are about 40,000 more business units now operating throughout the country across a variety of industries.
Clearly all of that will have helped to fuel the demand for property in some way. But each segment of the property market is different and each has its own drivers. So let me move on and make some remarks about the major areas of property - rural, commercial and industrial and housing.
Prices of various rural land types - dairy, fattening land, etc - have tended in the past to move in a similar fashion. This suggests that a similar set of factors is driving the demand in each sector. World growth, which in turn affects commodity prices, is probably one of these factors. Another reason why the prices of different land types tend to move together may be competition for land between different primary sub-sectors. For example, increased demand for dairy products will tend to increase the value of land used for dairying, but the effects are also likely to felt by other sub-sectors, lifting the price of land that is currently being used for other purposes.
Figure 1 shows changes in rural property prices and changes in the CPI. The profiles of the two series are clearly different. Consumer inflation was very high in the mid 1980s, but then fell to 2 per cent in late 1991. It has generally stayed low since then. In contrast, rural property prices showed large rises in the late 1980s and early 1990s before levelling off. Rural property prices have shown further gains in recent years, and increased by 12.9 per cent in calendar year 2003.
Figure 1. Rural property prices and the CPI
Annual percent change
What caused the large rises in prices that began in the late 1980s and continued over the 1990s? The levelling off in prices in 1991 appears to have been the result of a sharp fall in world commodity prices in the second half of 1990 on the back of a weak world economy.
The first thing to note is that rural property prices fell through much of the 1985-89 period, following the abolition of subsidies. In real terms, this fall was severe. In figure 1, we can view the gap between the rural property prices and the CPI as being the change in real property prices. The fall in real prices in the 1980s was huge, since nominal property prices were falling at the same time as consumer inflation was high. Given this, a lot of the rise in property prices that occurred in the early 1990s could be viewed as `catch up' with real prices correcting back to a more normal level.
However, other factors were also at work -- the exchange rate eased in late 1988 which flowed through into export prices; interest rates fell sharply between 1990 and 1992; a new government was elected in late 1990 and began carrying out further reforms; confidence was returning to farmers after they realised that they could in fact operate profitably without subsidies; and growth in the economy overall began to rise in early 1993.
Figure 2. Rural property prices, commodity prices and merchandise terms of trade
Indexes with base 1989Q4 = 1000
What has driven the rise in rural property prices since 2000? The initial boost came from an extraordinary rise in export earnings which occurred in the 18 month period beginning in June 1999. During this period world prices for our exports rose strongly, while the exchange rate declined. Figure 2, which shows property prices in index form (i.e. in level terms rather than growth rates), illustrates the strong climb in commodity prices as expressed in New Zealand dollars.
On top of this climb in export prices, export volumes also rose. Export earnings peaked in December 2000 and have declined a bit since then. Even so, they are still at a much higher level than they were in the early 1999. This is despite the current strength in our exchange rate (which, I might add, is due in part to the weakness of the US dollar). Fortunately our exchange rate appreciation has occurred at a time when world commodity prices have been high. Also, given the speed with which our currency has appreciated, some exporters still have significant foreign exchange cover in place. They had taken out much of this cover in the period when the exchange rate was low. This has partly offset the impact of the exchange rate rise on their earnings.
Another factor that drove land prices in the early 2000s was the conversion of farms to dairying. Over the second half of the 1990s, dairy prices rose relative to those for alternative pastoral products like meat and wool, with the positive effects of the Uruguay GATT round becoming apparent. The formation of Fonterra may have also been a factor in the move to dairy.
In the last two years, other forms of farming have come to the fore. The continuing fall in sheep numbers in the EU, and residual anxieties about BSE and CJD in Britain, have pushed lamb prices to new highs. At the same time, the BSE outbreak in the US, which was traced back to a Canadian herd, has resulted in North American beef being virtually shut out of the world market. With demand for beef from North Asia continuing to be strong, world beef prices have risen sharply.
While the outlook for agriculture remains positive, it is too early to say that prices for the commodities that we produce have shifted up a level and will stay there. It can be argued, for example, that increasing demand for dairy products from China means that our dairy prices will move to a higher level. Even if this was the case, it would be unwise to think that commodity price cycles would disappear altogether -- prices will continue to cycle, even if they cycle around a higher level.
The National Bank, in its Rural Report of March 2004, suggested that even now rural land might be too expensive. The National Bank notes that the value of an asset in economic terms is the present value of future expected income discounted at the required rate of return. The National Bank estimates that for the future income stream to equal the current price of rural land a discount rate of around 4 per cent is required, which is very low. Provided that the future income stream is being estimated correctly, this suggests that rural land is currently overvalued.
Analysis that we have undertaken at the Reserve Bank indicates that the ratio of rural land prices to agricultural operating surplus is now above its long run average value. However, the ratio is not yet out of line with the values that it reached in the mid 1990s.
Whether there will be a downward adjustment in prices presumably depends on whether market participants also reach the conclusion that rural land is overvalued. Rural dwellers often remind us that non-economic factors -- lifestyle considerations in particular -- are also important reasons for wanting to hold rural land.
Industrial and commercial property
The prices of industrial and commercial buildings rose sharply in the mid 1980s during the growth surge that followed the first moves to deregulate the economy. By industrial buildings I mean factories, cool-stores, warehouses and the like. By commercial buildings, I mean offices, retail buildings, hotels, and other similar places of business. Looking back, we can see that we had a price bubble. (Bubbles are often difficult to identify when they are occurring, but are clearly obvious once they've burst.)
In a bubble, asset prices become disconnected from reasonable expectations of the future earnings of those assets. Markets fail to get prices right. This mis-pricing gets reinforced and exaggerated by herd behaviour, or irrational exuberance. Investors convince themselves that someone else will pay even higher prices for the assets in future.
In the case of commercial buildings in the 1980s, the pace of construction was frantic, as supply rose to meet the high demand which was manifesting itself in high prices. Anyone who was around at that time can remember the cranes that cluttered the skylines of our major cities. While prices for industrial properties also rose sharply in the 1980s, construction of new industrial buildings was actually fairly steady during this period.
The bubble that occurred in the mid 1980s was not limited to business property. The SE40 share market index doubled in one year and then halved in the following year, after the crash. Nor was the bubble limited to New Zealand; it occurred in other countries too, notably in the US.
The bubble burst in late 1987 when the US share market crashed. It suddenly became obvious that asset prices had been out of line with economic fundamentals. We had witnessed a sustained period of misplaced investment, with the returns from this investment proving to be low. Misdirecting resources in this manner can be very costly for the economy.
The consequences for New Zealand were serious. Some companies went bankrupt and the economy went into a recession. It didn't recover from this recession until 1992. In terms of the loss of output relative to potential output, this recession was probably New Zealand's second worst of the twentieth century.
Figure 3. Industrial property prices and the CPI
Annual percent change
When the recovery did arrive, it was strong. Manufacturing, much of which had been restructured and was running under new ownership and management, began to thrive. For a number of years we had double digit percentage growth in manufactured export volumes. Consequently, the demand for industrial property rose sharply. In the newly deregulated environment of the labour market, employment growth was strong, and unemployment began a steady decline. The growth in service sector employment increased the demand for commercial property.
The Asian crisis and the drought of 1997 slowed demand for both industrial and commercial properties. But prices have lifted again in recent years, fuelled by the economic growth stemming from all the factors I mentioned earlier.
In general, these recent rises do not appear to be cause for concern. As I see it, we have learnt from the lessons of the past, especially the lessons from the 1980s. An incremental approach to industrial and commercial building appears to have been adopted. Little speculative building is being done, and arrangements regarding the tenancy of new buildings are often finalised before building begins. Additions and alterations continue to be a major component of total building activity, with office space being refurbished in order to meet clients' needs. The clients too have probably played their part, by moving to open plan arrangements and paring back their requirements for floor space. I hope these trends continue.
Figure 4. Commercial property prices and the CPI
Annual percent change
I do see some future challenges for the commercial and industrial property sectors. Building consent data over the past 12 months have indicated an increase in new building intentions in the sector. High levels of activity in the housing and apartment sectors have deflected some resources and labour away from the commercial building sector. Those pressures are likely to remain in the near future. In addition, planned government investment in roading and other areas of infrastructure will continue to place heavy demand on civil engineering and related professions as well as the demand for labourers, many of whom might otherwise choose to work in the property construction field.
In fact, there are currently pressures on factors of production across all sectors of the economy. For a number of years now, firms across all industries have been reporting that it is getting harder to recruit both skilled and unskilled labour. There appears to be an emerging view on the part of employers that, over the last 10 years or so, the country as a whole has not done enough training, particularly in the skilled trades area. Hence, delivering on commercial construction projects over the next few years will be a challenge requiring careful management.
I'd now like to make some comments about New Zealand's residential property market, which has experienced a strong cyclical upswing over the past three years. During that period, we have seen record numbers of house sales together with a significant lift in the construction of new dwellings, both houses and apartments. Unlike the residential upturn during the mid 1990s, which affected mainly the upper North Island, this one has been spread across the country, including many parts of the South Island.
House prices have increased substantially and by significantly more than we've seen in other New Zealand property markets over the same period. Indeed, Quotable Value New Zealand data suggest that house prices measured across the country as a whole have increased by nearly 50 per cent over the past three years. In some regions, the increase has been much more dramatic than that.
Figure 5. House prices and the CPI
Annual percent change
Most market observers, the Reserve Bank included, agree that the upswing has now peaked and that demand is gradually beginning to cool. House sales, which are a good barometer of demand and a good leading indicator of future building activity, have edged down over 2004. The number of new building consents issued, although fairly volatile, appears to be easing after rapid growth in both 2002 and 2003. We're also now seeing some cooling in the rate of growth in credit extended for housing purposes.
Figure 6. Indicators of housing activity
Annual percent change
Nevertheless, both house sales and new consents remain at high levels by historical standards and current residential construction activity is very high as the sector continues to work off a considerable backlog of demand built-up over the past couple of years.
To some degree, the residential construction sector has been able to enhance its own capacity to supply. The Household Labour Force Survey shows that employment in the wider construction sector has risen by nearly 40,000 people since 2001. A number of building companies have gone to great lengths to overcome shortages of labour either by accelerated training or by tapping into labour markets abroad. Even so, the sector remains stretched, with clear shortages of particular skills. The frustration households seem to face in finding a builder, plumber or other tradesperson at reasonable notice these days has become part of the national folklore.
In talking about the various property sectors, it's fair to say that economic analysts in New Zealand, including analysts at the Reserve Bank, tend to focus more heavily on the housing market than other markets. One of the reasons for this is simply that this is where much of the action has been over recent years; activity and prices in the markets for most other types of property, other than perhaps rural property, have been relatively more subdued. Another reason is the role that residential construction plays in economic growth. Residential construction accounts for around 6 per cent of total GDP, which is about twice the amount accounted for by non-residential construction. (However, if `other construction', which includes infrastructure spending, is added to non-residential construction, the amount gets closer to the residential construction total.)
Another reason why analysts are interested in the housing market is the `wealth effect'. A rise in house prices increases the wealth of households. In fact, in recent times, house prices have tended to be the major driver of changes in household wealth. A rise in household wealth in turn results in a rise in household consumption; with households feeling richer, they tend to spend more on consumption items. Given that private consumption accounts for nearly 60 per cent of expenditure on GDP, it can be seen why we take such an interest in the `wealth effect', and what house prices are doing.
Over the past three years, these linkages have been of particular interest to monetary policy. The upsurge in housing activity and construction has added directly to domestic inflation pressures. Residential construction costs, as measured by the Consumers Price Index, have increased by nearly 20 per cent, contributing significantly to overall inflation. We've also witnessed very strong household spending over this period which appears to have been reinforced by the rapid increase in house prices. Whilst I would not want to overplay the significance of housing and construction in our policy decisions -- stronger inflation pressures have been evident in many other parts of the domestic economy as well -- we have clearly had to take the strong housing sector into account when determining policy settings.
The recent period of strength in the residential property market is hardly unprecedented in New Zealand. The early 1970s, the early and late 1980s, and the mid 1990s were also periods marked by intense activity in the housing market and strong house price inflation.
There were some unique features to each of those cycles, but also some common drivers. Each coincided with a substantial acceleration in population growth to levels well above normal, due mainly to a spurt of high net immigration -- more arrivals and fewer departures. Each cycle was also reinforced by some other stimulus, such as a lift in export prices received from abroad, fuelling household incomes.
A sharp lift in net immigration and the sharp improvement in export returns from about 2000 through to 2002 were also catalysts for the recent upturn, although there are now indications that the migration pressures on housing are easing. Figure 7 shows estimates of the annual demand for dwellings from migrants. These estimates, which are indicative only, were derived by assuming that the number of persons per household would be the same for migrants as for the rest of the population. As figure 7 shows, the demand from short term visitors has been negative over the last year, as the number of visitors leaving the country has outnumbered those arriving. This has largely reflected a very sharp fall in the number of short-term overseas students in the country over the past 18 months (i.e. those here for periods of less than a year).
Figure 7. Estimated annual demand for dwellings from migrants
Number of dwellings (calculated as net migrants/average persons per dwelling)
There have also been other factors acting to reinforce demand for new dwellings in recent years. These include the drift of New Zealanders to warmer regions and into the cities, as well as social changes that have seen the average number of persons per dwelling steadily decline. Life-style changes and preferences have increased the demand for inner-city apartments and more exotic alternatives to the traditional New Zealand family home -- the one bathroom, three bedroom bungalow. Strong economic activity, which means more income and more jobs, gives households the capacity to accelerate these changes. But when the supply of such housing is inelastic -- as it always is in the short-run -- the result is upward pressure on house prices and construction costs.
Another source of demand during the latest cycle, at least in its early stages, has been the significant demand for properties by non-residents particularly in coastal and lakeside regions. The relatively low New Zealand dollar up until about 2002 helped to make such properties particularly attractive to foreign buyers. Although we have no reliable way of telling how much of New Zealand's housing stock is now owned by people living abroad, that proportion has almost certainly increased substantially over the past few years. Demand coming from people living abroad is likely to be less sensitive to monetary policy than demand coming from resident population.
The housing markets in some parts of the country where such activity was prevalent a year or two ago, such as Nelson, appear to have been cooling recently. The significant rise in house prices in these regions following a surge in demand, coupled with the stronger New Zealand dollar, has presumably dampened overseas investor enthusiasm to buy such properties. However, by all accounts, Australian investors are still active in purchasing New Zealand properties at the moment.
All of the factors I have mentioned help to explain why housing demand has been strong, but they are not the full story. One of the more noteworthy aspects of the housing upturn has been that very similar cycles have been seen in a large number of other countries around the world. Along with New Zealand, many countries, including Australia, the United States, United Kingdom, Ireland, France, Italy, Spain, and some other OECD countries, have all experienced very strong housing markets with significant increases in house prices in recent years. In many cases their upturns started a little earlier than New Zealand's and the subsequent cooling seems a little more advanced. An associated feature is that debt to income ratios have continued to lift sharply in many of these countries over recent years, including Australia, the United States and New Zealand, reflecting the enthusiasm for buying houses. However, the question to arise is what the common drivers, if any, might have been?
The investment motive seems to be a common factor. Households in most of these countries appear to have viewed investment in housing as a preferred alternative to other forms of savings and investment. In many countries, there also appears to have been something of an aversion on the part of the household sector to other forms of investment, such as shares or superannuation funds. That aversion is likely to reflect the losses that some investors incurred at the beginning of the decade as the world economy slowed and the `tech-wreck' unfolded. In our own case, one only has to look at the low level of net inflows into managed funds over the past few years, to see how investors have behaved in the wake of poor returns received earlier in the decade.
Relatively low interest rates in most countries in recent years have also undoubtedly made the debt financing of housing purchases relatively more attractive for many households. One might argue that interest rates in some countries were set at too low a level over this period, but it should be remembered that until quite recently central banks have had to contend with weakness in general activity in many of these countries, notwithstanding stronger housing markets.
Those buying a house primarily driven by an investment motive may or may not choose to live in the house themselves. An increasing number of purchases appear to have been by those wishing to let the house on the rental market and expecting to make a capital gain. We lack comprehensive statistics on such activity in New Zealand, but our contacts in the banking sector confirm that a substantial part of the recent growth in housing credit has been for that purpose. Investor housing activity has, of course, been a key driver of the recent property boom in places such as Sydney and Melbourne, as the Reserve Bank of Australia has noted. Such activity often relies on a steady stream of rental income in order to meet the financing obligations on the property.
Growth in housing rentals in New Zealand has been lagging rising house prices for some time now, and thus rental yields in many parts of the country appear to be declining. Consequently, the success of `housing as an investment' may largely depend on the prospect for sustained capital gain over the coming years. Last year I commented on the potential vulnerabilities that some investors could face when the housing market or the economy inevitably cools.
Those vulnerabilities arise either from being disappointed in respect to capital gain or being unable to meet outgoings should interest rates rise further or the rental market weaken in the future.
Looking at the balance sheet of New Zealand households one might well ask whether these vulnerabilities are overstated. The recent sharp rise in house prices has to date made New Zealand households considerably more wealthy, at least on paper.
To illustrate that proposition, the Reserve Bank's own estimate of the household sector's net wealth (including the current market value of housing) stood at $345 billion (about 3 times annual GDP) at the end of 2003. That was up from $260 billion (or about 2.5 times annual GDP) in 2001. This improvement in net wealth was despite households taking on an extra $23 billion worth of debt over the same period. Surely, we would need a very large and unprecedented fall in house prices to reverse that improvement?
Figure 8. Net wealth of households
$billion as at December
The answer, of course, is that the aggregate household balance sheet gives very limited perspective on the exposures of individual households or investors. For example, some households or investors are clearly considerably more highly geared than the average New Zealand household represented in the balance sheet figures I just quoted. Moreover, there is a composition issue here. New Zealanders hold a very large, and increasing, portion of their wealth in housing. That itself creates a potential vulnerability. Past experience shows that individual house prices can and do fall by significant amounts even if the national average house price appears comparatively resilient.
Consequently the Reserve Bank has been giving a consistent message to households and investors over the last year. Prudent buyers and investors need to satisfy themselves that they could withstand a reasonably significant fall in house prices and rentals and/or a reasonably significant rise in interest rates. In housing, as with any other investment, it's the investor who takes the risk, thus it's the investor who needs to be careful.
I should point out that the Reserve Bank is certainly not projecting a calamitous fall in house prices over the next few years. However, some of the fundamental drivers of the housing cycle that I mentioned before, such as rapid population growth, certainly appear to be easing, and the evidence does point to a cooling market. A reasonable view is that house prices are unlikely to rise much further over the next two years, and some falls are certainly possible, particularly in some regions.
Financial System Stability
Although the possibility of falls in house prices at some point in the future is something investors in housing need to be wary about, the Reserve Bank is also interested in what the consequences of a widespread fall would be for the stability of the banking system. More generally, we are also interested in the potential stability implications of a significant change in values for other types of property such as commercial or rural properties, against which the banking system extends significant amounts of debt.
Last year, New Zealand participated in the International Monetary Fund's (IMF) Financial Stability Assessment Programme (FSAP). In preparation for the FSAP, the Reserve Bank, in conjunction with the major banks, examined the potential vulnerability of the banking system to a significant fall in house prices combined with a marked rise in unemployment.
One aspect of this exercise was to look hypothetically at what might occur if house prices did fall substantially and if the unemployment rate increased sharply, given current lending exposures. The exercise assumed movements that were extreme, but by no means implausible, by international standards.
I am pleased to say that the results of this stress test were favourable -- the banking system itself appears well placed to withstand a marked fall in house prices and an associated deterioration in the labour market should these events ever occur. In part, this reflects measures banks have taken to effectively insure themselves against the risk of default on housing lending. Of course, on matters related to financial stability there's never room for complacency. Moreover, this positive finding does not remove the onus on individual households and investors to be careful. While the result of the stress test does give us a measure of confidence in the likely resilience of the banking system to a marked fall in house prices, some individuals could nevertheless be hurt if such a scenario was to eventuate.
As part of the same exercise, we also examined the possible effects of a sizeable fall in both commercial property prices and corporate earnings for the banking system. Once again, this exercise suggested the banking system is well placed to absorb such a shock. And again, this positive finding does not remove the onus on commercial property investors to exercise appropriate care as they go about their business.
Summing up, property market developments in New Zealand over the past few years can be explained largely in terms of the economic cycle. The relative strength of many segments of property -- in terms of prices, rentals, vacancies, or new building activity --largely reflects growing levels of demand for property as the scale of the economy and the number of people in it expands. To that extent, these developments provide little reason for concern.
However, like any other asset markets, property markets can get out of kilter with the underlying requirements of the economy and investor preferences can change independently of the economy at large. As I have said before, at the margin this may have been the case in parts of the housing market over the past two years, with some investors becoming unrealistic about prospective returns. There are, no doubt, examples of overzealous investors in the commercial and rural property markets too. Since property markets, and the economy that they serve, are inherently subject to cycles, market participants need to remain wary of the risks and structure their affairs accordingly.