Face - Index - Top/Bottom - Previous/Next

"Monetary review - 4th Quarter 2000"

Continued from previous page

Property prices

In the short to medium term property prices are to a large extent determined by demand, which follows the business cycle, especially interest-rate levels (financing conditions) and the households' real incomes ^[15]. Chart 5 presents annual rates of increase in real cash prices and a measure of the cyclical situation – the output gap. This measure is calculated as the deviation of actual real output from an estimate of potential real output.

Chart 5 Real cash price and output gap in Denmark

Property prices also depend on the general price level, the property stock and tax regulations (e.g. the tax value of the deductibility of interest payments).

In the short term at least the supply of properties is relatively inelastic. An increase in the demand for housing compared to the housing stock exerts upward pressure on cash prices. New construction thus becomes more attractive, and housing investments will increase, so that the housing stock will gradually expand. In the long run housing prices are determined by construction costs.

In certain periods housing prices may deviate significantly from the general development in prices, cf. Chart 6. Since 1993 increases in housing prices have been considerably greater than increases in construction costs. However, this acceleration occurred after a period of sluggish housing prices at the end of the 1980s and the beginning of the 1990s. Compared to construction costs, the cash price index is close to the relatively high level seen in the mid-1980s, cf. Chart 7.

Chart 6 Cash prices, consumer prices and construction costs in Denmark

Chart 7 Cash price as a ratio of construction costs

The correlation between property prices in various countries is considerably lower than the correlation between stock prices. The interest-rate parity and the effects of the international business cycle have an intensified impact on the stock market due to such factors as the opportunities for international arbitrage and the very liquid nature of the stock market. The property market is influenced first and foremost by domestic conditions. Hence, the capability of economic policy to affect the property market is greater than its capability to affect the stock market.

Households have a fundamental need for housing, but not for a portfolio of stocks. Prices for housing thus tend to be less forward-looking than stock prices. An investor in stocks may choose other liquid investment alternatives, such as bank deposits, in certain periods if a lower return on stocks is expected in the immediate future. For a home owner, the alternative is to rent a home. However, moving in and out of the market for owner-occupied homes is cost-intensive, and furthermore the market for rented housing in Denmark is inefficient.

Asset prices and financial activity

Strong fluctuations in asset prices have a significant impact on the solvency of the financial sector and the course of the economy. The most frequently cited effect of asset price fluctuations on the business cycle is the effect from the private sector's wealth to shifts in consumption and investments (wealth effect) ^[16]. For the corporate sector rising asset prices will reduce the cost of acquiring new capital compared to existing capital, which can increase the level of investment (the Tobin's q effect). Asset prices are generally forward-looking and rising asset prices may reflect expectations of a higher level of future economic activity. Expectations of higher future income will increase consumption and investments (expectations channel) ^[17]. Finally, effects related to the functioning of the credit markets (credit channels) also play an important role. A case in point is that it is easier for financially sound households or business enterprises to borrow, and at a lower interest rate (lower external financing premium), than would have been the case if their financial position were less sound (balance-sheet effect). Assets are used as collateral to enhance access to borrowing and to reduce the premium on external financing, and thereby cushion the effects of adverse selection ^[18]. Strong fluctuations in asset prices will thus affect the private sector's access to credit. For example, if the market value of the assets of the private sector declines, while payments on liabilities remain unchanged, creditworthiness is reduced, and thereby also opportunities to achieve the desired level of credit financing. The external financing premium will also tend to increase. The banks are subject to capital adequacy requirements, i.e. they have an incentive to reduce the scope of lending if the value of the collateral declines. This type of credit rationing will affect small business enterprises and households in particular, since their dependence on bank loans is relatively high. The mechanism, called a financial accelerator, amplifies the fluctuations in credit extension and thereby also reinforces business cycles ^[19].

The financial sector is exposed both directly and indirectly to fluctuations in asset prices. The direct exposure is via capital gains and losses on its own asset portfolios, while the indirect exposure is via the portfolio of lending to the business sector and households. Falling asset prices undermine the solvency of the private sector and may cause the number of defaulted loans to rise. This diminishes the financial soundness of the banks and reduces opportunities to offer new loans (lending channel). The effect may be amplified if the banks have to realise the assets provided as collateral for the loans, such as stocks, at a time when the asset price is falling sharply. This may again adversely affect stock prices and the banks' solvency, resulting in a vicious circle which can lead to a credit crunch. The opposite effect can be seen when asset prices are rising strongly, leading to a credit boom.

The principal asset of households is owner-occupied homes ^[20]. A very large proportion of Danish households own their own home, and this proportion is rising. In 1999 approximately 61 per cent of all households owned their own home, an increase from approximately 47 per cent in 1970. In most OECD countries the proportion of owner-occupied homes is more than 50 per cent. There is a relatively close correlation between the development in housing prices and the lending by banks and mortgage-credit institutes in Denmark, cf. Chart 8.

Chart 8 Real housing prices and lending by banks and mortgage-credit institutes

Most of the lending to households by banks and mortgage-credit institutes is to finance homes or to finance consumption backed by real property mortgaged as collateral ^[21]. A high marginal tax rate, together with the tax deductibility of interest payments, provides an incentive to borrow against the free mortgageable value in step with rising property prices. Unless the loan is reinvested in the home, borrowing against the free mortgageable value makes households more vulnerable to falling property prices. New credit products have furthermore affected the borrowers' exposure to changes in interest rates. For example, falling housing prices due to an increase in interest rates will also reduce the debt if the home is financed by a long-term fixed-rate loan. This reduction in the market value of the debt does not equivalently apply to variable-rate loans.

The households' stock portfolios accounted for less than 5 per cent of net wealth in Germany, France and Italy in 1997, and thus the significance of stock prices to credit granting is less pronounced in Europe. In the USA, on the other hand, the stock portfolio constituted almost 25 per cent of net wealth in 1997, so that credit granting in the USA is more sensitive to stock-price fluctuations, cf. Chart 9.

Chart 9 Real stock index and bank loans as a ratio of GDP in the USA, 1970-2000

In the USA maximum 50 per cent of an investment in stocks may be financed by borrowing from a stockbroker (margin borrowing). During the 1990s the growth in margin borrowing remained relatively close to the growth in stock-market value, but with a tendency for stronger increases in the last two years.

The financial markets tend to grow faster than the overall economy. Between 1985 and 1998 the value of outstanding bank loans, bonds and shares rose from around 150 per cent of GDP to around 250 per cent in the major OECD countries. In Denmark a similar trend was seen. In view of the appreciation of households' assets as a ratio of GDP, attributable in particular to the strong increases in asset prices, but also households' greater participation in the financial markets, fluctuations in asset prices can be expected to have a stronger impact on the general economic situation.

Denmark currently has a sound financial sector and a well-consolidated business sector. This is typical of a period with high asset prices ^[22]. The households also show relatively sound balance sheets. The interest burden fell during the 1990s, but the households are still vulnerable to fluctuations in property prices. The mortgage ratio, i.e. lending in mortgage-credit bonds as a ratio of the property value, was relatively stable at around 65 per cent during the 1990s, but somewhat higher than in the mid-1980s when mortgage-credit lending was still subject to a number of restrictions.

Monetary policy

The primary objective of monetary policy in most OCED countries is to maintain price stability. Stable asset prices are not an objective per se. However, large fluctuations in asset prices affect financial and economic stability and can thus have an indirect impact on the primary objective. As a consequence, central banks pay due attention to asset prices in their planning of monetary policy.

Asset prices play an important role since over- or underestimation can amplify cyclical fluctuations, to the detriment of price stability and long-term economic growth and employment. Furthermore, speculative bubbles can distort the utilisation of resources in the economy. For example, over-estimation of an asset will reduce its capital cost and lead to excess investment in that asset. This leads to inefficient allocation of resources, which is comparable to the effect of consumer price inflation.

Asset prices are also important because they contain information on the expectations of future economic activity and inflation which is not contained in other variables. Since assets are claims on future consumption they can in theory be interpreted as the price of future consumption and can therefore be applied to the overall assessment of the future course of inflation. The impact of monetary policy on the real economy is subject to a long time lag, typically more than one year. Forward-looking information from the asset markets can therefore be relevant to the conduct of monetary policy. Financial variables are particularly relevant to economic forecasting, due to their instantaneous nature. Furthermore, they are not subject to revisions and also involve forward contracts, which gives an opportunity for more long-term forecasts than would otherwise be the case.

In Denmark monetary policy is designed to fulfil the fixed-exchange-rate policy, which excludes use of the interest-rate instrument to counter the effect of large fluctuations in asset prices. Other OECD countries such as the USA and the euro area use the interest-rate instrument to influence the economy's course, in order to achieve the primary objective. By adjusting the official interest rate the central bank's monetary policy can influence demand from households and business enterprises for goods and capital goods, thereby ensuring stable price development ^[23]. However, monetary policy can also affect economic activity indirectly via the impact of interest-rate adjustments on asset prices, e.g. stock prices and housing prices. For example, reducing the interest rate will increase the discount factor and thereby also asset values. The impact on asset prices may affect the behaviour of households and business enterprises, since fluctuations in asset values affect the size of the net assets (the wealth effect) and may change the access to borrow and the external financing premium by influencing the value of the collateral (the balance-sheet effect).

A key issue is whether central banks should use the interest-rate instrument to counter a speculative bubble in asset prices with a potential impact on the primary objective of the central bank.

A traditional argument against is that when a central bank focuses solely on its primary objective of a stable consumer price index it also implicitly takes the development in asset prices into account. If the markets for goods and service are efficient, the forward-looking information from asset prices will be contained in current prices in the markets for goods and services. It will therefore be sufficient to conduct monetary policy according to the consumer price index ^[24]. However, efficient markets for goods and services is a strong assumption to make. If the assumption does not hold, the central bank can gain more information on the economy's future course by also including asset prices.

The inclusion of asset prices in a total cost index together with consumer prices, and the application of this index to the conduct of monetary policy, may enable the central bank to implement more timely measures, with smaller interest-rate adjustments than would otherwise be required. Should a speculative bubble in an asset arise, it could be appropriate to raise the official interest rate, even though this will brake economic activity in the short term, and push inflation down. Timely measures would make it possible to avoid a subsequent strong decline in activity and inflation, and achieve more stable development in output, employment and inflation in the long run ^[25].

Including asset prices in monetary-policy planning can also pose significant problems, however.

Firstly, the impact of asset prices on economic activity and inflation may vary over time. Periodic fluctuations in asset prices can be caused by different underlying factors and have varying overall impacts on real variables.

Secondly, there may be a considerable level of noise in asset prices, which reduces the reliability of the information.

Thirdly, there is a risk of reduced transparency in monetary policy. The individual objectives may send different signals, requiring different responses. In the event of conflicting objectives it can be difficult to communicate simple reasons for monetary-policy adjustments. If there is doubt about the priority given by the central bank to various objectives the credibility of monetary policy will be reduced and its effects weakened.

Fourthly, it is difficult for a central bank to have an explicit or implicit asset-price target. The current value depends on future variables which are subject to great uncertainty. As previously stated, it is not possible to observe the fundamental discount factor, making it difficult to determine whether the current prices include a non-fundamental component.

Even if a central bank is able to determine that asset prices are not fundamentally justified, it is not certain that the central bank's policy instrument is the most appropriate. A speculative bubble in a specific asset makes it desirable to adjust the relative price between the asset and other goods and services, but an interest-rate adjustment has a broadbased impact on the economy. In certain situations other elements of economic policy can have a more effective impact, e.g. via supervision of financial corporations, accounting regulations, and financial legislation, and the elimination of some of the distorting effects of the tax system.

In Denmark the primary objective of monetary policy is to ensure a stable krone rate, while the stabilisation of business cycles is an element of fiscal and structural policy. The economic-policy response to the strong increases in housing prices and private consumption during the 1990s is an example of the application of the fiscal-policy instrument. One of the intentions of the Whitsun package of economic measures was to stabilise the property market by dampening growth in property prices to a level equivalent to inflation ^[26]. Rising housing prices in the mid-1990s led to expansion of private consumption, partly due to borrowing against free mortgageable property values. The savings ratio fell, and a current-account deficit accumulated. The ensuing tightening of fiscal policy included a reduction of the value of the tax deductibility of interest payments. Since then the propensity to consume has declined and growth in housing prices has subsided. Reducing the value of the tax deductibility of interest payments can also contribute to reducing future fluctuations in asset prices as it provides an incentive to reduce speculative gearing, i.e. to limit speculation based on borrowed funds.

The primary objective of the Eurosystem is consumer-price stability. The monetary-policy strategy to achieve this objective is based on two pillars. One is the development in the monetary aggregate, M3, and the other is the development in a wide range of economic and financial variables which are of relevance to the development in consumer prices. The development in asset values such as stock and housing prices is an element of the second pillar, and is therefore monitored on an ongoing basis. Furthermore, a speculative bubble in an asset tends to evolve simultaneously with an expansion of the monetary aggregates. The first pillar can therefore indirectly provide signals of asset price trends, and strong increases in the money stock may provoke a monetary-policy response.

There is an inherent risk of a more immediate economic policy response to declining than to increasing asset prices. Strong asset price fluctuations typically have a stronger impact on the real economy than gradual fluctuations, even though the two trends can be of similar size in overall terms. Increases in asset prices are typically gradual, while decreases can be very abrupt in some cases. This may therefore indicate an asymmetrical economic-policy response whereby investors perceive an automatic safety net under falling asset prices and "benign neglect" in the case of rising asset prices. The impression of an asymmetrical response is an inappropriate signal which may induce investors to increase their risk exposure, e.g. by accepting a lower risk premium in stock investments (moral hazard). If major increases in asset prices are followed by strong expansion of lending by banks, increases in the investment ratio of the private sector, a falling savings ratio and considerable current-account deficits, the overall signal will be that the development is not sustainable. It may be important to demonstrate a willingness to act, should several indicators point to an overheating.

Footnotes

[15] A cash price relation is estimated in e.g. Lone Schøtt Jensen and Dan Knudsen, Housing Investments and Cash Prices (in Danish), Nationaløkonomisk Tidsskrift, 1990. The effect on cash prices of fluctuation in bond yields is described in The Monetary Policy Transmission Mechanism, Chapter 4 in Monetary Policy in Denmark (in Danish), Danmarks Nationalbank, 1999.

[16] The transmission channels from changes in asset prices to the real economy are described in Erik Haller Pedersen, Capital Gains on Stocks and Owner-Occupied Homes, Danmarks Nationalbank, Monetary Review, 4th Quarter 1998.

[17] An interesting correlation between the development in stock prices and consumer confidence in the USA is examined in Maria Ward Otoo, Consumer Sentiment and the Stock Market, Working Paper, Finance and Economics Discussion Series, no. 60, Board of Governors of the Federal Reserve System, 1999.

[18] Adverse selection applies when the lender is unable to gain a full insight on the borrower's ability to repay a loan. In view of the risk of an excessively large proportion of defaulting customers the lender may choose to reduce the volume of lending.

[19] The accelerating effect is described in e.g. Ben Bernanke, Mark Gertler and Simon Gilchrist, The Financial Accelerator and the Flight to Quality, The Review of Economics and Statistics, February 1996.

[20] Gross housing assets were approximately 290 per cent of disposable income in Denmark in 1999, while the portfolio of listed shares and mutual fund certificates was approximately 22 per cent. Housing makes up approximately 60 per cent of the households' total wealth, cf. Economic Survey, (in Danish) Ministry of Economic Affairs, May 2000.

[21] Danish households account for the largest mortgage debt in the EU at approximately 70 per cent of GDP. However, since the various countries' home financing structures differ considerably, this factor does not accurately express the households' total debt compared to other countries.

[22] A current assessment of the financial stability in Denmark is presented in Financial Stability, Danmarks Nationalbank, Monetary Review, 2nd Quarter 2000.

[23] It is normally assumed that the behaviour of households and business enterprises depends on real interest rates rather than nominal official interest rates. However, the nominal rigidity of prices and wages enables the central bank to affect short-term real interest rates by adjusting the official interest rates.

[24] This conclusion is reached in e.g. Ben Bernanke and Mark Gertler, Monetary Policy and Asset Prices Volatility, in New Challenges For Monetary Policy, Federal Reserve Bank of Kansas City, 1999.

[25] The result in Stephen Cecchetti, Hans Genberg, John Lipsky and Sushil Wadhwani, Asset Prices and Central Bank Policy, CEPR, 2000.

[26] See e.g. Families and Income, (in Danish) Økonomisk Tema, Ministry of Economic Affairs, November 2000.

Face - Index - Top/Bottom - Previous/Next

Version 1.0 December 2000 Nationalbanken.
Published by Danmarks Nationalbank December 2000, http://www.nationalbanken.dk