According to the Case-Shiller national index, US home prices peaked in the second quarter of 2006 and had fallen by 10% by last year’s final quarter. By contrast, the Office of Housing Enterprise Oversight (OFHEO) purchase-only index was essentially unchanged over the same period. Why the divergence?

Both indices are based on a repeat-sales methodology. They use different underlying data – deed records of residential sales transactions in the case of Case-Shiller and mortgages purchased by Fannie Mae and Freddie Mac in the case of the OFHEO – but the most likely explanation for the divergence is that the Case-Shiller index is value-weighted, so is more affected by price movements of expensive housing.

This difference in construction implies that Case-Shiller should be a better guide to changes in aggregate housing wealth but OFHEO is more representative of the price experience of the typical home-owner.

The chart below compares the two indices over a longer time-frame. Recent weakness in the Case-Shiller index is the counterpart of much greater strength during the boom period. Case-Shiller climbed 69% in the five years to the second quarter of 2006, while OFHEO was up 47%. Even after its recent hefty drop, Case-Shiller is still ahead of OFHEO since the latter’s inception in 1991.

Bears argue that the loss of wealth implied by Case-Shiller will prompt significant consumer retrenchment. If OFHEO is correct in suggesting the average home-owner has so far experienced only a modest decline, however, the effect could be smaller than feared. OFHEO would also suggest a lower estimate of the number of households currently in negative equity.

The Budget was a non-event in macroeconomic terms. The Chancellor’s strategy is to allow borrowing to take the strain of a weaker economy this year in the hope that growth rebounds solidly in 2009 and beyond. This may well be appropriate but claims that the fiscal rules are still being met look increasingly untenable.

Key points:

1. The Budget is broadly revenue neutral in 2008/9 but raises £1.9 billion by 2010/11, mostly via increases in alcohol, fuel and vehicle excise duties.

2. Public sector net borrowing is now projected at £43 billion in 2008/9, up from £36 billion in the Pre-Budget Report, reflecting the impact of slower growth and financial market turmoil on tax receipts.

3. General government net borrowing is forecast at 3.2% of GDP in 2008/9 – above the 3% Maastricht treaty limit.

4. Mid-point GDP growth projections have been revised down by a quarter of a percentage point in 2008 and 2009, to 2.0% and 2.5% respectively. 2010 is unchanged at 2.75%.

5. As usual, the claim that the Golden Rule will be met over the cycle relies on a projected sharp improvement beyond the coming fiscal year, based on a rise in the tax share of GDP as well as faster economic growth.

6. One surprise is that net gilt issuance is projected to surge from £29 billion in 2007/8 to £63 billion in 2008/9. As well as higher net borrowing, this reflects a transfer of the Northern Rock loan from the Bank of England to the Treasury and repayment of the Bank’s historical “Ways and Means Advance” to the government.

7. The transfer of the Northern Rock loan is required to comply with EU restrictions on central bank lending to government entities. The loan is projected to fall to £14 billion by March 2009, implying a reduction of £11 billion from its current estimated level of £25 billion. It is striking that even the Treasury, which has every reason to be conservative, projects a significant repayment by early next year.

Northern Rock could be in a position to repay most its loan from the Bank of England by early next year. Here’s how.

First, consider the Granite securitisation vehicle. According to documentation on Rock’s website, on 31 March 2007 Granite held £16 billion worth of fixed-rate mortgages with resets occurring in 2008. Rock is offering unattractive rates to refinance, while its standard variable rate is a high 7.59%. Any borrowers able to switch to other mortgage lenders are likely to do so. Let’s assume £13 billion of Granite mortgages are repaid in 2008.

Granite expects to repay principal of £9 billion on its outstanding notes in 2008. If £13 billion flows back from mortgages, this leaves a surplus of £4 billion. My understanding is that Northern Rock is able to extract this surplus by injecting mortgages from its own book into the Granite pool.

Rock’s non-Granite mortgages totalled £38 billion on 30 June 2007 so there would seem to be no obstacle to extracting surplus cash from Granite, even if new mortgage business is negligible, as seems likely. Moreover, a portion of these non-Granite mortgages will also be repaid this year. Assuming the same profile as for the Granite pool, £11 billion could be fixed-rate deals resetting in 2008. Let’s say £9 billion of non-Granite mortgages are repaid this year. Adding this to the £4 billion Granite surplus gives a total inflow of £13 billion.

Now consider funding. Rock had £24 billion of retail deposits on 30 June 2007 but at least half left the bank after it was forced to seek emergency funding from the Bank of England. Post-nationalisation, savings are returning in response to high interest rates and the unlimited government guarantee. (Savers were previously deterred by the risk of accounts being frozen if the bank entered administration.) Even after a recent cut, Rock’s tracker online and silver savings accounts offer a highly competitive 6.25%. If this edge is maintained, retail inflows of £10 billion or more look possible this year. (Landsbanki’s ICESAVE attracted £5 billion in the 15 months to 31 December 2007 from a standing start and without the benefit of a government guarantee.)

Adding a £10 billion deposit inflow to mortgage repayments of £13 billion would leave Rock only £2 billion short of the estimated £25 billion Bank of England loan.

What could go wrong with this scenario? One risk is that other lenders will be unable to accommodate borrowers switching from Rock given the current difficult funding environment. Perhaps this partly explains government measures announced last week designed to restart the market for mortgage securities.

Similarly, concerns about unfair competition or the stability of other institutions relying on retail funding may force Rock to cut its deposit rates, implying a smaller savings inflow.

Rock could also decide not to pay the Bank of England back so soon, even if it has the resources. It may wish to continue to generate new mortgage business, albeit on a much smaller scale than in recent years, in order to maximise its attraction to an eventual purchaser. Also, the Bank loan represents cheap funding, at least when considered from the perspective of the public sector as a whole. The Bank of England effectively pays Bank rate on the money it creates to lend to Rock. At 5.25%, this is a full percentage point below the rates Rock currently offers on its leading savings products.

Friday’s weak payrolls numbers are widely viewed as confirming that the US economy is in a recession. I think they are consistent with a flat economy. A recession may be coming but it has yet to be confirmed.

A recessionary labour market is characterised by a fall in hiring and a rise in layoffs. So far only the former has occurred. Layoff announcements have been stable in recent months – see chart. This is why weekly initial claims for unemployment insurance have yet to reach the 400,000 level that would signal a recession.

Interestingly, Manpower chairman and chief executive Jeffrey A. Joerres made the same point when commenting on his firm’s latest hiring survey, showing the weakest US jobs outlook for four years. “The important change is not about reductions in workforces, like we would expect in a recessionary period, but rather an increase in the percentages of employers who are planning to put a hold on hiring.”

According to the payrolls report, aggregate weekly hours worked in the private nonfarm economy contracted at a 1.7% annualised rate in January / February from the fourth quarter. This could still be consistent with stable or expanding output, given that productivity has recently been growing at a 2% pace.

The worsening credit crisis and soaring food and energy costs have increased near-term risks but there is substantial monetary and fiscal stimulus in the pipeline. The economy may yet muddle through. Wait for layoffs and initial claims to rise before accepting recession orthodoxy.

An earlier post argued that damage to US financial institutions from the subprime crisis could be similar in scale to the losses suffered as a result of the savings and loan crisis of the late 1980s. If so, the fall in financial stocks in 1989-90 could be a guide to the extent of the current bear market.

At the time of the last post the comparison suggested the bear market would extend in both price and time but a bottom might be in place by March, to be followed by a strong rally over the remainder of the year. The chart below provides an update. Financial stocks would have to fall by a further 10% to match the cumulative decline during the S&L bear market but are now in the time window for a low.

Bears argue that valuations are not yet as depressed as at the October 1990 trough. For example, the price to book ratio of the financials index is 1.46 versus 1.02 then. Relative to other sectors, however, the position is more extreme now: the price to book discount to non-financials is 52% versus 45% at the 1990 low.

Another bear argument is that loan delinquencies will rise further but that was also true when the October 1990 rally began: delinquencies peaked in the second half of 1991, by which time financial stocks had rallied by more than 60%.

The real level of the Fed funds rate is lower now than at the 1990 trough, while the Treasury yield curve is steeper. Credit spreads are still widening but they were at the time of the 1990 rally too, peaking only three months later.

Perhaps subprime losses will exceed S&L damage. If not, financials are starting to look attractive, particularly relative to other sectors.