Housing market developments are critical to the macroeconomic outlook. Although the level of housing investment is small relative to national output in most countries, it is very volatile and changes in housing investment can significantly affect the economy’s cyclical course. Perhaps more important, wealth in the form of housing equity is a large part of all wealth, often equal to 40% or more of total personal wealth. Housing inflation or deflation affects consumption, because of the resulting ‘wealth effects’. Finally, mortgage credit is also typically 40% or more of the banking system’s assets, and the rate of growth of mortgage credit therefore impacts on the rates of growth of bank assets and banks’ deposit liabilities (i.e., the quantity of money). Signs of a recovery in the US housing market are of great importance to the macroeconomic prospect in 2013. They matter partly because of the ‘wealth effects’ for consumption and the implications for housing investment, as mentioned above. But they also matter because of the likely strengthening of bank balance sheets, as delinquent loans are reinstated and mortgage credit starts to expand again.
The latest Federal Reserve data on delinquency rates on US banks’ assets, for the second quarter of 2012, show only a glacial rate of improvement in asset quality. In fact, the delinquency rate for banks’ residential real estate loans rose in Q2, even as delinquency rates for most categories of bank loan fell. But the news of house price increases has come more definitely in recent months (i.e., Q3). In September last year I sent out a bullish weekly e-mail on the American banking system, against the consensus at that time. I would like to repeat the general message in this weekly e-mail and to warn that, if US broad money starts to grow at 5% plus in 2013 (as seems plausible), the Fed would be very unwise to leave its funds rate at zero. Also in 2013 there will be a widening contrast between a reviving USA on the one hand, and seemingly chronic macroeconomic malaises in both the Eurozone and Japan.
US housing market on the turn
The American housing market has improved in 2012, with quite marked increases in house prices being reported in recent months. The S & P/Case Shiller index of urban house prices advanced by 0.9% in June and by a further 0.4% in July, and all the signs are that a similar rate of progress was seen in August and September. If the rate of increase in the two months of June and July continued for a year, that would given an annual increase of just over 8%. The chart below is not of the S & P/Case Shiller index, but of a separate index prepared by the Office of Federal Housing Enterprise Oversight. At any rate, it shows that the average rate of increase in the OFHEO home price index since 1991 has been just above 3% a year, while even in the first half of 2012 the annualized rate of increase (at 5.5%) was above this. Certainly, if the second half of 2012 sees a higher rate of house price appreciation than the first, the conclusion can be drawn that the US housing market has turned upwards.
Other signs of recovery: increasing turnover in existing stock
Changes in house prices matter in numerous ways to the macroeconomic outlook. For decades American economists have watched closely both new housing starts, a series which is well-known to have leading-indicator properties, and new housing permits, a series which might be a good leading indicator for starts if it were not buffeted about so much by the weather and other erratic influences. The latest information is that new starts in August rose slightly from July, to a seasonally adjusted annual rate of 750,000 homes and apartments. Applications for building permits fell slightly in August, but July had a value of 811,000 and was a four-year high.
These numbers may sound impressive, but a sense of perspective comes from noting that the all-time peak level of housing starts was in January 2006 at 2,270,000. In other words, US new housing construction now runs at a third of its highest-ever figure, less than seven years ago. Gross residential private fixed investment fell in real terms, according to US GDP data, by 7.3% in 2006, 18.7% in 2007, 23.9% in 2008, 22.4% in 2009, 3.7% in 2010 and 1.4% in 2011. (No kidding. The total decline from the 2005 level of residential construction was 57.5%.) That is an eloquent metric of the severity of the Great Recession that the USA has undergone; it is also a reminder of how much more the American economy could produce were demand to revive, and a hint of the strength of the potential recovery in store in 2013 and 2014. Residential construction today accounts for less than 2 ½% of US GDP, but – if it were to return to past glories – it could by itself add about 3% to national output.
It is important to understand that purchases of new homes are small, in the USA as in most economies, relative to transactions in existing homes. Nevertheless, the markets in new and existing homes are of course interrelated. The most recent peak in existing home sales in 2005, of just under 7.1m. units, coincided with the peak in gross private residential construction. Existing home sales have been rising in the last three years, but remain well below the 2005 figure. According to the National Association of Realtors, existing home sales in August were 4.8 million, 9.3% up on a year earlier. The growth of existing home sales has implications for a number of other macroeconomic variables. There is an obvious correlation between the level of housing turnover and the incomes of the professionals involved (i.e., the realtors themselves [or ‘estate agents’ in UK parlance] and the lawyers handling the conveyancing). But, more generally, people moving house tend afterwards to invest heavily in home refurbishment and the purchase of consumer durables. Sales of consumer durables are also a recognised leading indicator of wider economic activity.
A heartening development for the future is that home buyers have been sufficiently active in the last few quarters that the backlog of unsold homes has started to fall. In 2010 realtors had on average an estimated 9.4 months of unsold homes in their inventory (i.e., on their websites, notice boards, etc.). Last year this had declined to 8.2 months and in recent months it has dropped to just over six months. The National Association of Realtors also has its own house price index, which relates – logically enough – to existing rather than newly-built homes. According to this index, the prices of existing homes in the USA as a whole were 9.5% higher in August 2012 than a year earlier. In the west of the USA the house price gain in the year was as much as 16.3%. (See the Appendix, on buoyant Arizona housing.) Set against the background of very low interest rates, with zero Fed funds rates allowing the fixed-rate residential mortgage rate to fall to about 4% for much of 2012, a roughly 10% rate of house price inflation is striking. The Federal Reserve’s announcement of so-called ‘QE3’ included large official purchases ($40b. a month) of mortgage-backed securities. Since then the annual cost of a fixed-rate 30-year mortgage has tumbled to just over 3.6%, the lowest figure for decades.
Favourable knock-on effects for the banking system?
Continued rises in house prices are likely to have favourable knock-on effects on the US banking system. The higher the level of house prices, the better is the quality of any portfolio of mortgage loans. Simply, if the average loan-to-value ratio in a mortgage portfolio were 85% with a house price at 100, it is under 57% for a house price index of 150. (Of course, in the USA the important role of the Federal mortgage agencies – Fannie Mae and Freddie Mac – in providing mortgage guarantees complicates the discussion. There is no time here to take this aspect of the subject forward.) Indeed, a self-reinforcing and potentially unstable spiral needs to be mentioned. As house prices recover, banks write back bad debts and operate with stronger capital ratios. The stronger capital ratios allow them to expand their balance sheets more rapidly. The implied faster rate of growth of money balances is associated with rising asset prices, including the prices of residential real estate. That allows further writing-back of bad debts. And so on. If expectations of housing inflation are adaptive, that can add a further twist to the spiral. Suppose that the house price change in the last two years dominates expectations of the house price change in the next two years, which is not at first glance a silly idea. (It is a silly idea, as house price expectations should be determined by the current level of house prices relative to long-run norms, as interpreted in terms of variables of known importance, such as the house-price-to-income ratio. But the world is a strange place.) At the present rate of recovery, two years of, say, 5% plus of annual US house price inflation are likely to be recorded in early 2014. If someone borrows at 3½% to finance the acquisition of homes (for rent, in the easiest case), then the expected return on borrowing is the net rental yield (which we may take as 5%, for illustration) plus the house price appreciation of, say, 7½%. Then the expected annual total return is over 12½%, against an interest cost of 3½%, arguing for more incurring of mortgage debt to expand the number of homes held by the landlord. Heavy borrowing from the banking system creates more money, which further enhances house price inflation and expectations of future house price increases. Irving Fisher’s downward debt deflation spiral is notorious. But upward debt inflation spirals – of the kind seen in the USA between 2004 and 2006, and then so savagely reversed – are also possible.
It has to be conceded that – at present – there is no evidence of an upward debt inflation spiral emerging. The housing downturn of 2007 – 11 was of appalling intensity, much the worst since the Great Depression of the early 1930s. It has left a legacy of bad debts which will take several more quarters of balance-sheet recuperation to fix. Above is a chart of the delinquency rate on banks’ residential mortgage assets, with Q2 2012 as the last value. The positive news is that the rate is lower than at the peak and relatively stable; the bad news is that the rate has not fallen in recent quarters, despite the early signs of house price inflation. Perhaps – if house prices continue to move forward – the delinquency rate on this category of bank assets will start to return to normal. More hopefully, banks are charging off bad mortgage loans at a much slower pace than in 2009 and 2010. (Loans can be delinquent, in that borrowers are not paying interest or in some other way not complying with loan terms, but banks may still regard the loan principal as fully recoverable. Even when a loan has been ‘charged’, i.e., charged to the profit and loss account, it may be recovered, in whole or in part at a later date. The finer details in this subject of delinquencies and charge-offs are quite theological. The Federal Reserve website says that, ‘Charge-offs are the value of loans and leases removed from the books and charged against loss reserves. Charge-off rates are annualized, net of recoveries.’ Meanwhile, in its words, ‘Delinquent loans and leases are those past due thirty days or more and still accruing interest as well as those in nonaccrual status.’)
Summing-up: could another foolish housing boom develop in 2013 and 2014?
American monetary policy-making lacks firm intellectual moorings at present. The Fed’s large holdings of mortgage securities, and its so-called ‘QE3’ announcement that it will be an aggressive purchaser of more such securities in the next few quarters, are an indulgence in credit allocation. But – as Robert Hetzel has convincingly argued in his recent book on The Great Recession: Market Failure or Policy Failure? – the Fed has no mandate to interfere in credit allocation, since that is properly for private-sector decision-takers subject to market pressures. Bernanke is a creditist, not a monetarist. His intellectual bias has affected many key initiatives in the six years of his Fed chairmanship. Although numerous signs of a reviving housing market have begun to emerge, the Fed under Bernanke’s direction has opted to push mortgage rates down to the lowest levels since the 1960s. Top Fed officials are not interested in the growth rate of the quantity of money, broadly- defined. Is there a chance that the housing recovery in 2013 and 2014 could become so strong that mortgage credit again grows rapidly and, along with a resurgence in other forms of credit, is associated with, say, M3 growth of 10% or more? If so, the Fed’s pledge to keep its fund rates at zero until 2015 will prove to have been very unwise.
My view is that another foolish housing boom – a lurch towards the excesses of 2005 and 2006 again – is unlikely, but not to be ruled out as impossible. For the time being the banks still have a significant proportion of non-performing or delinquent mortgage assets on their balance sheets, and they will be reluctant to expand their mortgage lending at a rapid rate. US commercial banks’ total residential mortgage assets (revolving home equity loans, closed-end residential loans and mortgage-backed securities) stood at $3,407.7b. in February 2012 and at $3,461.3b. in August 2012, with an implied annualized growth rate of 3.2%. (Total bank assets at August 2012 were $12,855.1b., giving a residential mortgage asset proportion of just under 27%.) The latest drop in the mortgage rate ought to be followed by continued house price increases in the second half of 2012 and in early 2013 at an annual rate in the 5% – 10% vicinity, which will give extremely good returns to people who can find credit facilities to acquire residential real estate. The delinquency and charge-off rates on mortgage portfolios will decline, and banks already have strong capital/asset ratios by past standards.
It is all too plausible that by spring 2013 banks will be growing their mortgage assets by at least ½% a month, in the context of a general revival in private sector bank credit. If the Fed perseveres with the QE3 operations, the quantity of money could well be growing at an annualized rate of 5% – 10%. At that point the Fed will have either to drop QE3 (and perhaps reverse it, with sales of government and mortgage-backed securities) or to raise Fed funds rate to 1% or 2%. On one point I am confident. If the 5% plus annualized rate of house price inflation seen so far in 2012 continues through 2013 and 2014, the Fed will have to raise its funds rate to well above zero to prevent another house price bubble. There was no need for Bernanke to pledge a zero Fed funds rate until 2015, and he will come to regret his words.Tags: BBVA Compass, Capital One, Deutsche Bank, European Union, Federal Reserve System, Goldman Sachs, Huntington Bancshares, Jpmorgan Chase, United Kingdom withdrawal from the European Union, United States