Saturday, April 29, 2006

The word from IBM is SOA!

Wednesday night I went along to the NY WebSphere Users Group to hear about IBM's latest strategies, and the word on that is...



The presentation was pretty light on technical details (just the usual sales pitch about how implementing the massive WebSphere SOA stack - WebSphere Application Server, WebSphere Business Integrator, WebSphere MQ, etc, etc - is easy and will solve all your problems), but what was interesting was some of the common practices of SOA implementations that IBM are pushing:

  • Governance. When you want to automate a process part of the automation exercise involves understanding and mapping out your business processes. This can be a challenge to those who have only ever followed the process manually.
  • Reuse. Find services that need to be used by multiple business units within your business and implement them as common, shared services instead of every business unit implementing the same service.
  • Pushing the work to the customers. If you're going to expose something as a service inside your company, why not go that extra step and expose it externally to your customers? Let them do the work instead of having to employ customer service people to do it for the customer!



So aren't all of these practices just good practice anyway, not specific to SOA? Yes! If you were to implement any of the workflow tools on the market you would have to define your business process anyway. It will save you money to find common services across your business or push work to your customers anyway.

However, if you can use the banner of SOA to justify implementing good practices then I'm all for it!

And that's the word. (Apologies to Stephen Colbert.)

Wednesday, April 26, 2006

Financial Ramifications of a Housing Slowdown

I'm currently doing a Certificate in Finance at NYU. For this semester's report I researched some of the interesting things happening in the US housing market and looked at what might happen if there is a slowdown (many people are expecting one in the second half of 2006).

Table of Contents

Table of Contents. 1

Introduction. 1

Measures of a Housing Bubble. 1

Anatomy of a Housing Bubble. 2

Analysis of the US Housing Market 3

Negative Savings. 3

Risky Loans. 3

Mortgage Equity Withdrawal 4

Flippers 4

Securitized Mortgages. 4

Derivatives. 5

Impacts of a Slowdown. 6

Conclusion. 7

Reference. 8

Introduction

This paper examines some of the risks and possible ramifications of a slowdown in the US Housing Market in 2006.

Measures of a Housing Bubble

What is a housing bubble? One definition of an asset bubble is: “If the reason that the price is high today is only because investors believe that the selling price is high tomorrow - when ‘fundamental’ factors do not seem to justify such a price – then a bubble exists. At least in the short run, the high price of the asset is merited, because it yields a return (capital gain plus dividend) equal to that on alternative assets.” The ‘dividend’ portion of the return from owning a house comes from the rent the owner saves by living in the house rent-free and the capital gains come from house price appreciation over time. A housing bubble can be thought of as driven by homebuyers who are willing to pay inflated prices for houses today because they expect unrealistically high housing appreciation in the future [Himmerberg, p. 68].

Three major measures are commonly used to assess whether house prices are unsustainably high: house price growth, price-to-rent ratio and price-to-income ratio.

A widely-quoted measure of housing prices is the Office of Federal Housing Enterprise Oversight (OFHEO), a single-family house price index. This index provides data on over 100 metropolitan areas since 1980. A major drawback is that the index only covers so-called “conventional” mortgages that have been purchased by Fannie Mae and Freddie Mac, which in 2005 has a maximum limit of $359,650 and therefore high-priced houses are underrepresented in the OFHEO index.

Price-to-rent ratio is a similar measure to a price-to-earnings multiple for stocks. It is intended to reflect the relative cost of owning vs renting. In theory, when house prices are too high relative to rents, potential homebuyers will instead choose to rent, thus reducing the demand for houses and causing house prices fall until they are once more in line with rents. If this ratio remains high for a prolonged period then it must be because prices are being sustained by unrealistic expectations of future price gains instead of fundamental rental value. i.e. bubble.

Anatomy of a Housing Bubble

Historically, the price-to-rent ratio for the nation as a whole has had many ups and downs, but over time it has tended to return to its long-run average. Thus, when the price-to-rent ratio is high, housing prices tend to grow more slowly or fall for a time, and when the ratio is low, prices tend to rise more rapidly.

Currently the ratio for the country is higher than at any time since data became available in 1970 - about 32% above its long-run average. Of course, the results vary widely from place to place. For Los Angeles and San Francisco, the price-to-rent ratio is about 40% higher than the normal level, while for Cleveland the ratio is very near its historical average.

The fact that the ratios for the nation and many areas of the country are higher than normal doesn’t necessarily prove that there’s a bubble. House prices could be high for some good reasons that affect their fundamental value. The most obvious reason is the low level of mortgage rates. This stems both from very stimulative monetary policy, which I discuss in more detail below.

The fact remains, however, housing prices have risen very fast for a historically long sustained period of time in many countries, it is easier to get credit than at other times and there is a glut of investment worldwide.

According to estimates by THE ECONOMIST, the total value of residential property in developed economies rose by more than $30 trillion over the past five years, to over $70 trillion, an increase equivalent to 100% of those countries' combined GDPs. Not only does this dwarf any previous house-price boom, it is larger than the global stock market bubble in the late 1990s (an increase over 5 years of 80% of GDP) or America's stock market bubble in the late 1920s (55% of GDP). In other words, it has the potential to be the biggest bubble in history. [ICW]

The global boom in house prices has been driven largely by historically low interest rates and low returns in the equities market in recent years.

Analysis of the US Housing Market

In recent years the percentage of home ownership in the US has been increasing. In 2004 68% of households owned their own homes and, for most of them, housing equity will make up nearly all of their non-pension assets at retirement. Many of the 32% who rent are younger households. [Himmelberg]

Negative Savings

At the end of 2005 the US personal savings rate fell into negative territory at minus 0.5%, the lowest point since the Great Depression. That means that people not only spent all of their after-tax income last year but had to dip into previous savings or increase their borrowing. Soaring home prices – which have been increasing at an average of 12% in recent years - apparently have convinced people they don't have to worry about saving, a belief that could be seriously tested as 78 million baby boomers begin to retire.

The savings rate has been negative for an entire year only twice before - in 1932 and 1933 - two years when Americans had to deplete savings to cope with the massive job layoffs and business failures caused by the Great Depression.

This time Americans are spending all their incomes and more because they feel wealthier because of soaring value of their homes, which for many Americans is the largest investment they own. A rebound in stock prices after the 2000 market collapse has also increased consumer confidence.

Risky Loans

A ‘traditional’ housing loan could be defined historically as 20% down payment and a fixed interest rate.

There’s been a huge rise in subprime, zero-down and interest-only loans in recent years. People who have had trouble getting loans in past decades because of poor credit or lack of a down payment are now eagerly sought after, while those with more solid finances are urged to stretch further and further to buy ever more lavish homes. This could be putting some real estate markets more at risk than they’ve ever been.

New, riskier forms of mortgage finance also allow buyers to borrow more money than they would have in the past. According to the NAR, 42% of all first-time buyers and 25% of all buyers made no down-payment on their home purchase in 2004.

Nearly 9% of the mortgages made in 2003 were 'subprime' (made to people with troubled credit or uncertain finances). That’s up from 4.5% in 1994. In terms of volume, $388 billion in subprime-mortgage loans were originated in the first nine months of 2004 - more than triple the amount made in the eight years before that.

Zero-down loans totaled more than $80 billion in 2003. They were virtually nonexistent a decade ago. Another relatively recent phenomenon is the 125% loan. It allows people to borrow more than their homes are worth.

A record number of homeowners have taken out Adjustable Rate Mortgages (ARMs). From 2001-2003 ARMs made up fewer than 20% of all new mortgages, a rate that is lower than all but one year in the 1990s. However, ARMs rose to 34% of new mortgages in 2004. [Himmelberg] ARMs expose consumers to more risk, because their payments can rise along with interest rates.

Interest-only mortgages require buyers to make only interest payments for a period of time. After that, the rate adjusts, requiring principal payments as well. The loans were designed for people who wanted to preserve their cash flow for other investments or who receive a lot of their compensation in lump sums, such as big annual bonuses. Interest-only loans were, until 2000, almost exclusively a product for the wealthy, who had plenty of real estate exposure in their portfolio. Now they’re being pushed as a viable alternative for the average homebuyer - who is probably underestimating the potential risk he’s taking on. In 2004 about 18% of the $1.8 trillion in securitized mortgages were interest-only.

Mortgage Equity Withdrawal

A major trend to emerge in recent years is for home owners to extract equity from their existing homes and use it for additional consumer spending. The equity is extracted via cash-out refinancing, home equity borrowing and housing turnover. Goldman Sachs [GS137] refers to this as Mortgage Equity Withdrawal (MEW) and they speculate that this extracted equity has a statistically significant and large effect on consumer spending. Between 50% and 62% of such equity is used for consumer spending, flowing into consumption or home improvement.

TODO: MEW spending is fine as long as house prices continue to rise. If there is a housing slowdown, however, MEW funds are very likely to dry up quickly.

Flippers

The inflated prices of real estate have given rise to the ‘flipper’ phenomena, which in turn inflates the real estate prices further. A study by the National Association of Realtors (NAR) found that 23% of all US houses bought in 2004 were for investment, not owner-occupation. Another 13% were bought as second homes. “Flippers” buy and sell new properties even before they are built in the hope of a large gain. In Miami as many as half of the original buyers resell new apartments in this way; many properties change hands two or three times before somebody moves in.[ICW]

Securitized Mortgages

Investment banks and other firms have been buying mortgage loans from lenders and packaging them into securities for sale to investors since the 1980s. But investor demand has surged in recent years, largely because in an era of low returns, mortgage-backed securities offer yield-starved investors much higher returns than government bonds.

U.S. lenders will make about $2.8 trillion in home-mortgage loans this year, according to the Mortgage Bankers Association (MBA). The MBA estimates that about 80% of these loans will end up in mortgage-backed securities. Mortgage-backed securities outstanding at the end of the first quarter totaled $4.61 trillion, up 61% since the end of 2000. In the same period, total Treasury securities outstanding grew 35% to $4.54 trillion. [REJ]

Investors' strong demand for mortgage debt, besides allowing lenders to offer many borrowers better terms, has also made it easier to offer mortgages to borrowers who might not easily qualify for a loan. The growth of the mortgage markets spreads the risks around. But some mortgage-industry analysts say lenders have become less stringent in their loan terms because they can sell almost any type of loan to those who package mortgage securities for investors.

In a world of low interest rates, the market for mortgage securities is simply too big and profitable for many investors to ignore. Investors can earn about 5.5% on mortgage securities whose payments are guaranteed by Fannie Mae or Freddie Mac, government-sponsored companies. Those who can stomach greater risk can buy subprime mortgage securities, which come with no guarantee but can yield as much as 15%, according to Bear Stearns. Typically pays 0.75 to 1.15 percentage points more than US treasuries.

The buyers of mortgage-backed securities include U.S. pension funds, hedge funds European insurance companies and - more recently – the central bank of China. But overseas investors are the fastest-growing source of demand. The trade publication Inside MBS & ABS estimates that foreigners held $280 billion of U.S. mortgage securities at the end of 2004, or 6% of the total outstanding. The foreigners' holdings rose 26% in 2004 and have continued to increase.

While Asian investors have largely focused on triple-A-rated bonds, other investors are buying lower-rated debt. These bonds, which are created when bankers carve up pools of mortgages, offer higher yields, but also bear the first risk of losses should borrowers default. Investors who buy these bonds in effect set the standards for which mortgages are made by deciding how much extra yield they need to compensate for the added risks of lower-quality loans. They include real-estate investment trusts, hedge funds and investors from Europe.

Strong investor interest has also made loans available to borrowers with poor credit and many other people who might otherwise have trouble getting a mortgage. Subprime loans included in mortgage securities totaled $401.5 billion last year, nearly double the total for 2003, according to Standard & Poor's. Meanwhile, loans with less than full documentation of the borrower's income and assets accounted for 70% of mortgage securities rated by Standard & Poor's in this year's first half, double the level recorded in 2000.

Derivatives

Also of concern is the amount of derivatives in "insured" commercial bank portfolios. The commercial banks, or course, do much of the lending to the housing sector. The amount increased by $2.6 trillion in the third quarter of 2005, to a whopping $98.8 trillion. 98% of these are concentrated in 5 banks. Total assets of these top 5 banks is $3.3 trillion. [OCC]

Most of the $570 trillion in derivatives are held by overseas investors.

Impacts of a Slowdown

After the biggest boom in US history the housing market has started to slow. Inventories of unsold homes are rising, homebuilding sentiment is falling and prices have fallen significantly in recent months.

The US housing bust in the 1990s decreased house prices by 25%, most of which occurred in the first 3 years, but prices did not start to recover for 5 years. In comparison the crash in Japan's housing market has been going on for 15 years and prices in some areas are down over 70%. Historically the US market fundamentals have been stronger than many overseas countries, with recovery from such bubbles taking between 5 and 7 years.

There are two ways for the housing bubble to ‘correct’ itself (i.e. return to normal trends): either incomes rise or housing prices drop - or a combination of both.

There are two important questions for the economy as a whole: Will the slowdown be gradual or will prices crash? And, how large will the impact on GDP be?

Goldman Sachs [GS137] predict the following:

At the national level house prices will stagnate in coming years. At mortgage rates of around 6% Goldman Sachs estimate prices are 15% overvalued, with a range of 0%-50% across regions. (Regions such as Boston, New York and parts of California are thought to be more overvalued.) Stagnation for a 3- to 5-year period would undo the overvaluation assuming median household incomes raise of 3%-5% per year.

Residential investment is expected to 15%-30% over the next couple of years. This is likely to trigger a significant drop in constructions employment, whose share of total payrolls is the highest in 50 years.

In terms of GDP the direct and indirect effects of a housing slowdown is expected to cut 1.5 percentage points from real GDP growth in 2007. Even allowing for a boost from net trade this could push growth from the above-trend 3.5% pace of the past two years to a below-trend 2.5%.

The direct hit to growth is likely to come from normalization in residential investment, which is likely to cut 0.75% from annualized GDP growth. The indirect hit will come from a sharp drop in Mortgage Equity Withdrawal (MEW), which is expected to cut a further 0.75% off annualized GDP.

Core inflation appears to be contained and should remain contained during a housing slowdown.

A Bernanke-controlled Federal Reserve Bank is expected to react to the slowdown by cutting its federal funds rate target by about 100 basis points in 2007.

Similarly, in order for the price-to-rent ratio to return to long-term norms, either rents must rise sharply or prices must fall. However, the Federal Reserve Bank cannot allow rents to rise sharply as this would feed into inflation. Rents directly or indirectly account for 29% of America’s consumer-price index, so raising inflation would force the Fed to raise interest rates more swiftly, which could trigger a fall in house prices. Alternatively, if rents continue to rise at their current annual pace of 2.5%, house prices would need to remain flat for over ten years to bring the US price-to-rent ratio back to its long-term norm. There is a clear risk prices might fall. [SWB]

What’s the likely effect if national house prices did fall by 25%, enough to bring the price-to-rent ratio back to its historical average? With housing wealth nearing $18 trillion today, such a drop in house prices would extinguish about $4.5 trillion of household wealth - equal to about 38 % of GDP. Standard estimates suggest that for each dollar of wealth lost, households tend to cut back on spending by around 3.5 cents. This amounts to a decrease in consumer spending of about 1.25% of GDP. To get some perspective on how big the effect would be, it’s worth comparing it with the stock market decline that began in early 2000. In that episode, the extinction of wealth was much greater - stock market wealth fell by $8.5 trillion from March 2000 to the end of 2002. This suggests that if house prices were to drop by 25%, the impact on the economy might be about half what it was when the stock market turned down a few years ago.

Wealth effects - positive or negative - tend to affect spending with fairly long lags. Therefore, a drop in house price probably would lead to a gradual cutback in spending, giving the Fed time to respond by lowering short-term interest rates and keeping the economy steady.

Faced with higher payments, homeowners who have drained all the equity in their home or who didn’t put anything down to begin with are much more likely to simply walk away. The same is true for those with troubled credit. The serious delinquency rate, where borrowers are 90 days overdue or in foreclosure, is 1 in 12 for the subprime market, versus about 1 in 100 for the prime. Delinquencies tend to peak two to three years after subprime loans are originated, said Glenn Costello, an analyst at Fitch Ratings in New York. Peak rates of about 20 percent to 25 percent now will likely rise to the high-20s in 2006, he said. [BLOOM]

So far, the mortgage-backed market has generated cash and profits galore. Someday, however, someone will lose money. No one knows who, when or how much because housing has never been so frothy, risky mortgages so common or the market in which they trade so vast. Also unknowable is the quality of lenders' and investors' hedges, the complex investing strategies that are supposed to cover losses in the event of turmoil, like a default-inducing spike in interest rates. Many of the players in this market are not subject to government regulation, and those that are haven't exactly been under the microscope.

Conclusion

While there is some debate about whether there is a US housing bubble or not, it is clear that housing prices have risen very fast for a historically long sustained period of time, it is easier to get credit than at other times and there is a glut of investment worldwide. Increasing risky lending practices and an as yet untested complicated web of securities and hedges (securitized mortgages, derivatives, etc.) has potentially set up an environment where a minor trigger event could set off a chain of defaults that will deepen and extend a slowdown.

However, US economic fundamentals appear to be strong and worldwide markets appear strong, so the likely outcome would be a slowdown on a similar scale to previous slowdowns – a 3-5 year dip with a 1.5% drop in GDP.

Reference

[BLOOM] Bloomberg.com, “Housing Bubble Bursts in U.S. Mortgage Bond Market”, December 6, 2005

http://quote.bloomberg.com/apps/news?pid=nifea&&sid=a2P0e3wlzcg8

[GS137]. Jan Hatzius. February 2006. Goldman Sachs Global Economics Paper No. 137. https://portal.gs.com

[Himmelberg] Himmelberg, Charles, Chrostopher Mayer and Todd Sinai. 2005. “Assessing High House Prices: Bubbles, Fundamentals and Misperceptions”. Journal of Economic Perspectives – Volume 19, Number 4 – Fall 2005

[ICW] The Economist, “In Come the Waves”, June 16th, 2005. http://www.economist.com/displaystory.cfm?story_id=4079027

[OCC] OCC Bank Derivatives Report Third Quarter 2005

Comptroller of the Currency Administrator of National Banks

http://www.occ.treas.gov/ftp/deriv/dq305.pdf

[REJ] RealEstateJournal.com, “Housing-Bubble Talk
Doesn't Scare Foreigners”, August 25, 2005

http://www.realestatejournal.com/buysell/mortgages/20050825-simon.html

[SWB] The Economist, “Still Want to Buy?”, March 3rd, 2005. http://www.economist.com/displaystory.cfm?story_id=3722894

Thursday, April 20, 2006

The Development Abstraction Layer

Joel talks about what it takes to build a successful software company, not just cutting code and hoping to somehow "market" it:

Any successful software company is going to consist of a thin layer of developers, creating software, spread across the top of a big abstract administrative organization... The abstraction exists solely to create the illusion that the daily activities of a programmer are all that it takes to create software products and bring them to market. Your first priority as the manager of a software team is building the development abstraction layer... If a programmer somewhere is worrying about a broken chair, or waiting on hold with Dell to order a new computer, the abstraction has sprung a leak.

Many of the same lessons apply to successful consulting engagements. If you are bringing in consultants to just cut code then you need to create an environment (manage) where they don't have to worry about getting accounts set up, finding the right person in a dependent team to talk to, having enough memory on the developer's PC, get a quick response for info when someone reports a production problem, etc. If you don't plan on creating such an abstraction then you better find someone else who will, otherwise the consultants will thrash between lots of tasks and the system stands a good chance of not getting delivered!

Wednesday, April 19, 2006

ExplorerCanvas - hack up Internet Explorer

Firefox, Safari and Opera 9 support the canvas tag to allow 2D command-based drawing. ExplorerCanvas brings the same functionality to Internet Explorer. To use, web developers only need to include a single script tag in their existing web pages.

After stagnating for many, many years we are starting to see new features added to the browser. When some browsers fall behind some clever developers find ways to add/hack that functionality. (Back story for ExplorerCanvas here.)

Thursday, April 13, 2006

Getting the SQL out of Hibernate

Hibernate is a popular Java Object-Relational mapping tool (as if you didn't already know that!) . For debugging purposes you may want to see the actual SQL, including parameter values, that Hibernate sends to the database. Out of the box you can enable the following log4j config:


<!-- Log the actual Hibernate Query Language (HQL) expression -->
<category name="org.hibernate.impl.SessionImpl">
<priority value="DEBUG"/>
</category>

<!-- Log the substitution parameters in the HQL expression -->
<category name="org.hibernate.engine.QueryParameters">
<priority value="DEBUG"/>
</category>


which will log something like the following:


[main] DEBUG org.hibernate.impl.SessionImpl - find: select count(*) from ClientDO cc where cc.clientID = ?
[main] DEBUG org.hibernate.engine.QueryParameters - parameters: [1786712]
[main] DEBUG org.hibernate.engine.QueryParameters - named parameters: {}


However, HQL is not the real SQL. For that you need to enable more logging:


<!-- Log the real SQL -->
<category name="org.hibernate.SQL">
<priority value="DEBUG"/>
</category>

<!-- Log the parameter substitutions in the SQL statement. Unfortunately logging org.hibernate gives a lot of other junk as well -->
<category name="org.hibernate">
<priority value="DEBUG"/>
</category>


which adds something like the following to your log output:


[main] DEBUG org.hibernate.SQL - select count(*) as col_0_0_ from instance_property instancepr0_ where (instancepr0_.value, instancepr0_.value_type)=?
[main] DEBUG org.hibernate.jdbc.AbstractBatcher - preparing statement
[main] DEBUG org.hibernate.loader.hql.QueryLoader - bindNamedParameters() /root/u2 -> namespacePath [1]
[main] DEBUG org.hibernate.type.StringType - binding '/root/u2' to parameter: 1
[main] DEBUG org.hibernate.type.IntegerType - binding '9' to parameter: 2
[main] DEBUG org.hibernate.jdbc.AbstractBatcher - about to close PreparedStatement (open PreparedStatements: 1, globally: 1)
[main] DEBUG org.hibernate.jdbc.AbstractBatcher - closing statement


What would be really useful would be the ability to log the SQL, including parameters, only if Hibernate throws an exception, insteading of logging all of this other stuff. I must look into implementing that...

Friday, April 07, 2006

Unsiloing in the workplace

An interesting new trend is emerging:

One of the hottest buzzwords in human resources now is "unsiloing," an ungainly bit of jargon that means this: Take upwardly mobile employees out of their straight-up-and-down functional "silos" and move them around to where their talents are needed, with an eye toward helping them cultivate new skills... "The trend will open up some interesting possibilities for people who want to get ahead, especially if they're between the ages of 35 and 55"

And why would managers go to the expense and effort to promote unsiloing?

"CEOs will rarely admit it in public, but privately they say they're worried about having enough qualified people to fill leadership roles as baby boomers retire," says Kochanski. "So more big companies now are actively analyzing where the gaps are likely to be, and which skills they need to be giving people the chance to develop right now."