This is the text of a talk given at the Ariyadh Development Authority Conference on Housing in Riyadh (11 April 2001).

Introduction
This paper outlines a theoretical model in which a major determinant of asset prices within an economy is the amount of credit creation undertaken by the commercial banking system. This feature can appear very clearly where created credit is directed into home mortgages, and where the asset price in question is therefore that of residential property. Whilst evidence for the theoretical model is drawn from the United Kingdom, the model can be applied to any economy that adopts a conventional banking system as a major source of finance for the purchase of residential property. The willingness of commercial banks to create credit is seen to be at least as important as real factors, such as demographic trends, in determining the performance of residential property prices.

Business Model of the Commercial Banking System
A widespread belief relating to the business of commercial banks is that a bank takes deposits of money from those in surplus and then lends this money to borrowers who need finance. The difference between the amount of interest paid by the bank to the depositor and the amount charged by the bank to the borrower then represents the bank's income.

However, when viewed as a single system, it can be seen that commercial banks create money out of nothing for subsequent lending at interest. With a documented monetary history of some 1000 years, England provides good evidence for this statement. Initially, it was the King alone who had the legal right to manufacture money. He would manufacture coins from gold or silver and decree them to be legal as payment for goods and services. But from the late 17th century onwards, the money creation process came to be shared by the early banks. They issued paper receipts in exchange for customers' deposits of gold and silver coins. The bankers encouraged their customers to use these paper receipts as money, and not to present them for repayment in coin at the bank. Gradually, the public came to accept the practice of paying for goods and services with the bankers' paper money. Accordingly, the banks turned to society as lenders of money. But they did not lend the coins that had been deposited with them by their customers. Instead they printed more paper receipts, which were by now regarded as 'money', and lent them at interest. The process of printing paper cost very little, and huge profits therefore resulted for the bankers.

In the modern day, commercial banks continue to create money out of nothing for the purpose of lending. The method that they use to create money is of course more sophisticated, relying upon a variety of tools such as cheques and account ledgers. As the following selection of graphs show, the amount of money created by the banking system (the IMF proxy measurement for M2 is shown) is far greater than the amount of money created by the state (M0). Society is paying interest to the commercial banks on all of the 'bank money' that has been manufactured out of nothing.Various comparisons of money creation by the state and by the commercial banking system in USA, UK, Japan, Turkey, Saudi Arabia; source: IMF Financial Statistics Yearbook, 2000

It seems that the Saudi authorities face a forecast shortage of money with which to finance housing in future years. It has therefore been suggested that Saudi commercial banks should be more widely allowed to 'mobilise Saudi savings' and supply them to house buyers through housing loans. But, as we have seen, the banks do not mobilise savings. They simply create money out of nothing and then lend it to borrowers at interest. The state could create this same amount of money interest-free, and thereby save borrowers immense amounts of interest. So why allow the banks to enter the market in the first place?

In the United Kingdom a large proportion of the money supply has been created by means of bank lending for house purchase. Figure 3 shows the scale of this lending. Other 'advanced' economies display similar orders of magnitude for housing loans in comparison to money supply. Fig. 3. Debt components of UK M4 money supply; source: IMF Financial Statistics Yearbook 2000, Economist 1994, Council of Mortgage Lenders 2000

Bank Lending and Asset Prices
Because the commercial banks make profit by charging interest on created money, there is a temptation for them to create as much money as possible. For every unit of money they create, an extra amount of interest can be charged. When too much money is created, the prices of assets, goods and services tend to rise. When the rate of money creation is reduced, price rises tend to decrease. Since banks create most of the modern money supply, and since they create money by making loans, we often see a strong relationship between changes in asset prices and changes in bank lending. Two graphs below relate changes in lending against, firstly, changes in the Financial Times All-share index and, secondly, changes in United Kingdom house prices. Fig. 4. Change in real UK total domestic debt against change in real FTSE All-share index ~ 1971 - 1997; source: IMF Financial Statistics Yearbook and UK Office for National Statistics. Fig. 5. Real house price change against real property lending change ~ UK 1971 - 1993; source: Gordon Pepper, Money Credit and Asset Prices, 1994

Impact of Interest on Project Evaluation
Since most modern money is manufactured as the balance sheet counterpart to an interest-bearing loan, money itself can be said to be 'interest-bearing'. The projects that are undertaken with interest-bearing money are very different to those that are undertaken with interest-free money. For example, we can establish a comparison between two different housing projects, one to build a high quality house that will last for 100 years at a cost of 1100 and the other to build a low quality house at a cost of 1000 that will last for only thirty years. Let us imagine that both houses can be rented out at 150 per year for their respective lifetimes. Discounted cash-flow analysis might then be undertaken to determine which of the two houses is more profitable to build. If an interest rate of 10% is used to evaluate this set of cash-flows, it is the low quality house that is indicated as the more profitable to build. This despite the fact that, for only 10% more in construction cost, a house that lasts more than three times as long could be built.

Willingness to Lend
If bank lending is a major factor in determining residential property prices, then what are the factors determining the amount of lending by commercial banks into the residential property sector? These factors fall into two classes:

• 'micro-factors' such as the amount of borrower's collateral, reliability of borrower's income stream, borrower's credit history, debt service as a percentage of borrower's income, location of property;
• 'macro-factors' such as the regulatory regime, forecast performance of property market prices, competition among lenders.

By measuring the ratio of average mortgage to average income, analysts may depict the willingness of lenders to lend. That a relationship exists between lending and prices, and that prices are more volatile than lending, is clear. Some commentators argue that the direction of the causation is unclear. Does lending increase because prices are increasing, or do prices increase because lending increases? The banking model proposed in this paper, in which banks create money by making loans, helps us to understand that the direction of causation is in fact from lending to prices and not the other way around. Fig. 7. Average loan to average earnings against average house price to average earnings ~ UK 1970 - 1993; source: Council of Mortgage Lenders 2000, Economist 1994, Office for National Statistics 2001

Affordability By measuring the proportion of household income spent on a mortgage, analysts provide a measure of affordability in the residential property sector, one that takes into account both house prices, average mortgage size and interest rates. The following graph shows that whilst the average mortgage as a proportion of average earnings decreased during 1979 - 1980 in the United Kingdom, affordability also decreased. This is explained by the increase in short term interest rates that occurred during the period, affecting the many mortgages under which repayments were linked to variable rates. Fig. 8. Proportion of household income spent on mortgage against average mortgage as a proportion of average male earnings ~ UK 1971 - 1998; source: Economist 1994, Social Trends 2000

When bank lending leads to increases in residential property prices, owner occupiers are forced to undertake a higher debt burden, other factors remaining equal. Government policy aimed at restraining rises in asset and consumer prices often resorts to the use of increased interest rates. This can increase average mortgage repayments as a percentage of average income and can lead directly to large scale home repossessions. Fig. 9. Home repossessions against UK T-bill rate ~ UK 1981 - 1999; source: Council of Mortgage Lenders 2000, Economist 1994, IMF Financial Statistics Yearbook 2000

turnover and completions
During a monetary expansion, a willing pool of debt-financed buyers offer increased prices to sellers and thereby encourage increased market turnover. Meanwhile, price increases sponsor increases in house completions by commercial agents. The opposite tendencies arise where price increases fade into price stability or where prices decrease, particularly so where sellers cannot easily cover their mortgage with proceeds from the sale of their home. In these cases, as in the United Kingdom during 1989-92, market turnover can fall dramatically. This volatility in levels of activity in the housing market is unhealthy for the construction industry and related service providers, and may result in a misallocation of resources at extremes of the cycle. Fig. 10. Change in average real house price against number of property transactions in England and Wales ~ 1971 - 1999 and fig. 11. Number of house completions by private enterprise against house price change ~ UK 1971 - 1993; source: Office for National Statistics 2000, Economist 1994, Council for Mortgage Lenders 2000

Conclusion
Homeowners who rely upon conventional bank mortgage finance must bear the burden of interest costs on money created out of nothing by the banking system. If money is to be created out of nothing, it is better created interest-free by an agency of the government of Saudi Arabia. This would achieve the desired aim of more affordable housing. More preferable is that communities fund their house financing requirements through mutual schemes without the money-creating intermediation of the banking system or the state. Both of these alternatives would tend towards greater stability in the housing market than a system in which bank lending played the primary role. They would also improve the quality and variety of output, since commercial activity would be undertaken with less regard to the dictates of discounted cash-flow analysis. It is my belief that if the commercial banks enter the Saudi property market to a substantial degree during coming years, then the tendency will be for 'boom and bust' features to arise. This risk would be particularly high during the early years after banking deregulation took place, as substantial increases in bank lending for property purchase began to occur.