Global liquidity drives house prices

By: Ansgar Belke, University of Essen, IZA Bonn and ECB Observer group

Global money balances and global house price inflation

The quite expansionary monetary policy of the G3 countries (Euro area, US and Japan) in connection with foreign exchange interventions by many Asian countries - especially China with its dollar reserves now standing at 1.5 billion - has contributed to a significant increase of global money balances during the last years. Surprisingly enough, the strong monetary dynamics has not been accompanied by a permanent rise in consumer prices so far. At the same time, however, large parts of OECD countries have experienced very sharp increases in asset prices, such as real estate or share prices. Between 2001 and 2006, for instance, house prices have strongly increased in the US (55%), the euro area (41%), Australia (59%), Canada (61%) and a number of further OECD countries. Notable exceptions are Japan where house prices stopped their 15-year fall not earlier than in 2007 and, of course, Germany with its huge East German real estate overcapacities. It cannot be ruled out a priori that this development also has some connection with the abundant liquidity that exists worldwide. Many observers even interpret the increase in asset prices as the result of liquidity spillovers to certain asset markets.

 

So far, the relationship between money growth, consumer prices and asset prices has been little studied in an international context. In our new paper we address this issue more deeply and investigate the extent and some specific macroeconomic impacts of global liquidity in order to identify its interactions with global consumer price and housing price inflation, as suggested by a number of authors. Our main idea is that different price elasticities of supply lead to differences in the dynamic pattern of price adjustment to a liquidity shock. While goods prices adjust slowly to changing (global) monetary conditions due to the cheap outflow of consumer goods from emerging markets, house prices react much faster since supply of house prices cannot be easily expanded and thus disequilibria on the housing market are generally balanced out by price changes. We come up with the conclusion that the liquidity the Western world has plenty of has contributed to a significant increase of house prices in recent years (and finally culminated in the subprime crisis!). 

 

The concept of global liquidity and the global perspective in general

The concept of "`global liquidity"' has attracted considerable attention in recent years, although the empirical literature regarding this topic is still quite scarce. The first studies in this field used different indices of liquidity in seven industrial countries to explore the dimension of the relationship between liquidity and asset returns. They find evidence that there are important common components in G7 money growth and that an increase in G7 money growth is consistent with higher G7 real stock returns and lower G7 real interest rates. As a stylized fact, periods of strong money growth are likely to turn into periods of high inflation, especially if they are accompanied by asset price inflation. Given the fact that at least the first and the third observation fit quite well with the situation observed on the world financial markets at least until spring 2007, this scenario has most probably contributed to the more recent positive trend of inflation rates observed in the second half of 2007, for instance, in the Euro area.

 

Not only with respect to global liquidity but also with an eye on global inflation performance, available evidence becomes increasingly stronger that the global instead of the national perspective is more important when monetary transmission mechanisms have to be identified and interpreted. For instance, current macro research has established that around seventy percent of the variance of the inflation rates of OECD countries can be traced back to a common “world” factor. Moreover, the same research finds empirical evidence in favour of a robust error-correction mechanism, meaning that deviations of national inflation from global inflation are corrected over time. Hence, inflation is to a large degree a global phenomenon.

 

Moreover, I would like to argue here that (a) the traditional way of modelling inflation is much too country-centred, (b) a global approach is by far more adequate and that (c) the importance of global factors has increased significantly more recently. One important global factor, for instance, is certainly represented by the mounting pressure enacted by the ever higher degree of competition on the international goods and labour markets - a phenomenon which has to be mainly ascribed to globalisation. It appears fair to say that it the globalisation process has significantly contributed to the decrease of inflation rates since the eighties (and that this puts the significance of the past contribution of central banks to price stability high on the research agenda again).

 

Considering the development of global liquidity over time, the question is often raised whether and to what extent global factors can be made responsible for it. Current macro research comes up with the stunning result that around fifty percent of the variance of a narrow monetary aggregate can be traced back to one common global factor. As one prominent example of such a global factor, for instance, the extremely lax monetary policy stance of the Bank of Japan during the last years should be mentioned here. It has been characterised by a significant accumulation of foreign reserves and by extremely low interest rates - at some time even approaching zero. By means of carry trades, financial investors took out loans in Japan which they invested in currencies with higher interest rates. Such kind of capital transactions should, of course, have an impact on the development of monetary aggregates far beyond the special case of Japan.

 

In addition, one should really focus on global instead of national liquidity since national monetary aggregates have become more difficult to interpret due to the huge increase of international capital flows. In the same vein, one could argue that global aggregates are likely to internalize cross-country movements in monetary aggregates - due to capital flows between the different regions - that may make the link between money and inflation and output more difficult to disentangle in the single country case. Moreover, in today's linked financial markets shifts in the money supply in one country may be absorbed by demand elsewhere, but simultaneous shifts in major economies may well have significant effects on worldwide goods price inflation.

 

Monetary policy and house prices – the connection

Nowadays, there is no dispute about the impact of house price developments on the macroeconomy and on the role of fundamental factors other than monetary policy for house price developments. Monetary policy driven house price booms may fuel consumer spending and thus, aggregate demand and inflation via balance sheet and credit-channel effects - more potential collateral meaning lower risk premia in this context via the now-famous Bernanke/Gertler financial accelerator framework. The improvement in lending conditions may in turn lead to an expansion in loans and thus higher money growth. The most direct link between housing prices and domestic demand might be construction activity and in particular the construction of houses (dwellings).

 

However, studies specifically dealing with the impacts of monetary policy on house prices are still quite scarce.  Some researchers just let the data speak and use a baseline New Keynesian model as a theoretical benchmark, consisting of a Phillips curve to describe the supply side of the economy and include property prices in the demand curve and show that this restores an empirically significant monetary transmission mechanism.

 

Others stress the user cost of capital as an important determinant of the demand for residential capital. In this context, lower interest rates in the wake of higher money growth should influence mortgage rates and thus by decreasing the user cost of capital should raise the demand for housing capital. A similar effect should work on the supply side where easier financing conditions tend to stimulate housing construction. However, most of them focus on the effects of interest changes on house price changes but do not explicitly refer to monetary aggregates. 

 

Empirical analysis

In our background paper, we investigate quarterly time series from 1984Q1 to 2006Q4 for the United States (US), the Euro area, Japan, United Kingdom (UK), Canada, South Korea, Australia, Switzerland, Sweden, Norway and Denmark, so that in our analysis 72.2% of the world GDP in 2006 and presumably a considerably larger share of global financial markets are represented. In the next step, we aggregate the country-specific series on a PPP basis to obtain global series. The pattern of consumer price inflation indicates the moderate inflation which started to emerge around the mid-90s and has persisted in the recent years of global excess liquidity. The growth of house prices has been increasingly distinctive not only in the last 5 years but also at the end of the 1980s, lending support to the popular hypothesis of asset price inflation in real estate. Interestingly, both of these periods were accompanied by a sharp rise in liquidity. Global short-term interest rates were at a historically low level from 2002 to 2005, since the monetary policy stance was extremely loose during this period.

 

Just in line with our theoretical predictions, consumer prices react only with significant time lags to a liquidity shock in our basic vector-autoregressive model (VAR) model without house prices. In case of a positive interest rate shock, the positive reaction of the consumer price level yields the "price puzzle" which often occurs in other VAR studies. The next step is to augment our baseline model with the global house price index. We find that house prices react relatively fast to a global liquidity shock. This supports the view that loose monetary policy and ample global liquidity have significantly contributed to the bull market in the real estate sector. By contrast, the reaction speed of consumer prices to the same shock is much lower, though clearly significant in the long run. Thus, both house prices and consumer prices appear to be determined by monetary conditions, although at different time horizons.

 

An increase of house prices leads to higher consumer prices, a result which should be mainly due to wealth effects or balance sheet effects. Furthermore, a positive house price shock raises liquidity which may be due to rising credit demand. More surprisingly, a house price shock causes a hike in interest rates - the latter representing more or less the main instrument of the central banks. However, it has not yet been generally acknowledged until the subprime crisis that monetary policy makers are reacting directly to house price developments! It is also interesting to see that the "price puzzle" disappears. This supports the view that house prices are essential for a reasonable performance of macro models and that otherwise an omitted variable bias might occur. Hence, we would like to argue that house prices and inflation expectations are correlated, since the lack of an inflation expectation variable is often supposed to be the reason for the existence of the "price puzzle".

 

Conclusions

Against the background of our results the still high level of global liquidity has to be interpreted as a threat for future stable and low inflation and financial stability. Since global excess liquidity is found to be an important determinant of house prices, there might be at least two implications for the adequate conduct of monetary policy. First, monetary policy has to be aware of different time lags in the transmission from money to different categories of prices. In particular, strong money growth might be a good indicator of emerging future bubbles in the real estate sector. However, this pattern should, on the contrary, also be taken into account when assessing the consequences of a slowing down or smooth reversal in global excess liquidity – among them the risks and options in the light of Bretton Woods II. Second, house prices might well serve as indicators of future inflationary pressures on goods markets.

 

 

Link to the background paper

Belke, Ansgar, Orth, Walter, Setzer, Ralph (2008): Global Liquidity and House Prices: A VAR Analysis for OECD Countries, March 18^th.

http://www.makro.uni-essen.de/pdf/global_liquidity_180308.pdf, and

www.ecb-observer.com

---