Coyote Blog

Via Kevin Drum, Mark S. Tucker and Kevin Toch make the argument, if I understand it right, that school districts and state education organizations simply don't have the expertise or the capacity to handle the changes required to meet the standards that are being applied by efforts like NCLB (they also argue the tests themselves suck, but I am not going to address that issue).  By the way, you know I'm going to get worked up when the title of an article is "The secret to making Bush's school reform law work? More bureaucrats"

...we need a long-term solution, which can only lie in building
the capacity of the states, districts, and schools to reach the kinds
of goals contemplated by the framers of NCLB. This is not a simple
matter, but a vast, man-to-the-Moon kind of challenge. It means finding
people with the data management experience to build and administer the
very complex systems called for by the law. It means recruiting experts
who can help create truly world class curriculum standards so that
teachers will know what they are supposed to teach and students will be
able to reach the standards. It means identifying and training
thousands of educators who have succeeded in improving their schools to
provide on-site assistance at other failing schools, and recruiting
still others who can take those schools over if the current staff
cannot or will not rise to the challenge. It means creating and
expanding networks of talent-laden organizations--universities, think
tanks, for-profit and non-profit school companies--that have the skill,
experience, and management capacity to turn around individual schools
and entire districts. And it means greatly strengthening the
capabilities of the agencies that will coordinate this massive effort:
state departments of education.

Wow!  It's hard to even know where to start, but I guess my first thought is : What the f*ck have public schools been doing in the last 100 years?  Why, after an absolutely enormous spending growth over the last several decades, do districts still not have the ability to create world class curricula?  Why don't teachers know what they are supposed to teach?   Why is the system so talent poor, despite a huge increase in the number of administrators with various advanced education degrees at all levels of the system?  It's as if the highway department announced today that they didn't have the ability to design roads.

The first and last resort of every technocrat is to complain that the system is great, if only the right "smart" people could get put in charge.  These folks are making this same argument yet again.  Our public schools are fine, if we could just get the right experts in charge. 

Bullshit.

The issue is not the lack of expertise.  The issue is one of incentives and senescence in the system itself.  In this context,  NCLB is completely off the mark.   I work with government employees all the time.  There is a very clear difference between the incentives they see and the incentives I see in the market.  For government employees, the biggest incentive is to avoid missing some bureaucratic check box.  They are much more concerned that they not be found later in some audit to have missed a procedure or a required approval authority than with actual performance or productivity.  NOT, I want to emphasize, because they are bad or misguided people, but because that is how their incentive system is set up.  Their actions are entirely rational in the context of their incentive structure, but the results are no less disastrous.

For example, government managers of recreation facilities get almost no credit for improving the customer experience, a metric my company lives and dies for.  I have seen a government park manager do a great job obtaining funds from private sources to add a new facility to their park that pleased guests, only to get criticized for having the slope of an access ramp be 1/4 degree off ADA standards and have a grievance filed by the union that park visitation had gone up, creating more work for the government employees.  I spent an evening having a beer with that manager, and you can bet they are never going to try to actually improve the customer experience again.  As another example, I went in to my government landlord last week and just blasted them for their lack of customer service focus, for the fact that they are blocking me from making improvements customers are begging for.  They yawned, gave me no response,  and handed me a notice that they were missing some of our water testing paperwork and please get it to them ASAP.

NCLB just gives government schools another government wammy to be managed and avoided.  The authors will probably get their wish, and huge bureaucracies will rise up to manage the numbers and reports without anything being done to really improve education.  The authors lament that the California state education department has "only" 1452 employees.  I have every confidence that this "problem" will soon get fixed by California, and the number will balloon up nicely, long before children see any better education.

A while back I wrote a plea to just let GM die.  I said:

A corporation has physical plant (like factories) and workers of
various skill levels who have productive potential.  These physical and
human assets are overlaid with what we generally shortcut as
"management" but which includes not just the actual humans currently
managing the company but the organization approach, the culture, the
management processes, its systems, the traditions, its contracts, its
unions, the intellectual property, etc. etc.  In fact, by calling all
this summed together "management", we falsely create the impression
that it can easily be changed out, by firing the overpaid bums and
getting new smarter guys.  This is not the case - Just ask Ross Perot.
You could fire the top 20 guys at GM and replace them all with the
consensus all-brilliant team and I still am not sure they could fix
it. 

All these management factors, from the managers themselves to
process to history to culture could better be called the corporate
DNA*.  And DNA is very hard to change.  Walmart may be freaking
brilliant at what they do, but demand that they change tomorrow to an
upscale retailer marketing fashion products to teenage girls, and I
don't think they would ever get there.  Its just too much change in the
DNA.  Yeah, you could hire some ex Merry-go-round** executives, but you
still have a culture aimed at big box low prices, a logistics system
and infrastructure aimed at doing same, absolutely no history or
knowledge of fashion, etc. etc.  I would bet you any amount of money I
could get to the GAP faster starting from scratch than starting from
Walmart.  For example, many folks (like me) greatly prefer Target over
Walmart because Target is a slightly nicer, more relaxing place to
shop.  And even this small difference may ultimately confound Walmart.
Even this very incremental need to add some aesthetics to their
experience may overtax their DNA....

Changing your DNA is tough.  It is sometimes possible, with the right
managers and a crisis mentality, to evolve DNA over a period of 20-30
years.  One could argue that GE did this, avoiding becoming an
old-industry dinosaur.  GM has had a 30 year window (dating from the
mid-seventies oil price rise and influx of imported cars) to make a
change, and it has not been enough.  GM's DNA was programmed to make
big, ugly (IMO) cars, and that is what it has continued to do.  If its
leaders were not able or willing to change its DNA over the last 30
years, no one, no matter how brilliant, is going to do it in the next
2-3.

I would say the exact same thing is true of public schools: Their DNA is senescent.  Most are the equivalent of alcoholics who keep falling off the wagon and keep asking for more chances.  At some point, you just have to give up.  At some point, it is easier to just start from scratch.  After 30 years of trying, Sears still can't change itself so there is Wal-Mart.  After 30 years of trying, GM still can't change itself so there is Toyota.  After 30 years of trying, United Airlines still can't change itself so there is Southwest.

The only difference in education is that the government has to date suppressed the emergence of Toyota and Wal-Mart and Southwest because, well, because it can.  I am sure that United Airlines would have liked to ban competition from Southwest, but it does not have the coercive power of government.  Fortunately, in most industries other than education, the public gets a choice of offerings, and companies that customers don't prefer tend to die.

It's time to give school choice a chance, and radically shift the incentives for public schools in a way that the government can't with bureaucracy-based programs like NCLB.  Some public schools will thrive, and many will die in favor of private options, but our kids will be far better off either way.  It's time to stop doubling down on failure.  It's time to stop giving the alcoholic one more chance.

Postscript:  One of the reasons that competition is important is in the very definition of "expertise."  An expert is someone who presumably has been succesful at a certain activity when others have been less so.  We call Herb Kelleher an expert on airlines and customer service because he designed a model that kicked everyone else's butt.  But would you have called him an expert in 1972, before Southwest took off?  Probably not.  He was just one of many voices with diverse, untested opinions of what would make a better airline.  What eventually made him an expert, and the others less so, is he went out and applied his ideas and they were succesful.

So the author's want to send more "expertise" to the schools.  OK, who are the experts?  Nearly every public school is using the same version of the same failed model.  Some succeed more than others, but these differences tend to be incremental rather than radical, like the difference between Sears and Montgomery Ward rather than between Sears and Wal-Mart (or even Amazon.com).  So how can you even know who the experts are within the same failed system, where no one is really allowed to go out and fully test their ideas in practice?  What happens, in reality, is that "experts" in education are the ones that can best enthrall academics and politicians and think tanks with grandiose or politically correct visions.  I would argue that as of this moment there are no experts in education in the US and we have no hope of identifying them until we let entrepreneurs go out and start testing various new models.

Via Kevin Drum, Mark S. Tucker and Kevin Toch make the argument, if I understand it right, that school districts and state education organizations simply don't have the expertise or the capacity to handle the changes required to meet the standards that are being applied by efforts like NCLB (they also argue the tests themselves suck, but I am not going to address that issue).  By the way, you know I'm going to get worked up when the title of an article is "The secret to making Bush's school reform law work? More bureaucrats"

...we need a long-term solution, which can only lie in building
the capacity of the states, districts, and schools to reach the kinds
of goals contemplated by the framers of NCLB. This is not a simple
matter, but a vast, man-to-the-Moon kind of challenge. It means finding
people with the data management experience to build and administer the
very complex systems called for by the law. It means recruiting experts
who can help create truly world class curriculum standards so that
teachers will know what they are supposed to teach and students will be
able to reach the standards. It means identifying and training
thousands of educators who have succeeded in improving their schools to
provide on-site assistance at other failing schools, and recruiting
still others who can take those schools over if the current staff
cannot or will not rise to the challenge. It means creating and
expanding networks of talent-laden organizations--universities, think
tanks, for-profit and non-profit school companies--that have the skill,
experience, and management capacity to turn around individual schools
and entire districts. And it means greatly strengthening the
capabilities of the agencies that will coordinate this massive effort:
state departments of education.

Wow!  It's hard to even know where to start, but I guess my first thought is : What the f*ck have public schools been doing in the last 100 years?  Why, after an absolutely enormous spending growth over the last several decades, do districts still not have the ability to create world class curricula?  Why don't teachers know what they are supposed to teach?   Why is the system so talent poor, despite a huge increase in the number of administrators with various advanced education degrees at all levels of the system?  It's as if the highway department announced today that they didn't have the ability to design roads.

The first and last resort of every technocrat is to complain that the system is great, if only the right "smart" people could get put in charge.  These folks are making this same argument yet again.  Our public schools are fine, if we could just get the right experts in charge. 

Bullshit.

The issue is not the lack of expertise.  The issue is one of incentives and senescence in the system itself.  In this context,  NCLB is completely off the mark.   I work with government employees all the time.  There is a very clear difference between the incentives they see and the incentives I see in the market.  For government employees, the biggest incentive is to avoid missing some bureaucratic check box.  They are much more concerned that they not be found later in some audit to have missed a procedure or a required approval authority than with actual performance or productivity.  NOT, I want to emphasize, because they are bad or misguided people, but because that is how their incentive system is set up.  Their actions are entirely rational in the context of their incentive structure, but the results are no less disastrous.

For example, government managers of recreation facilities get almost no credit for improving the customer experience, a metric my company lives and dies for.  I have seen a government park manager do a great job obtaining funds from private sources to add a new facility to their park that pleased guests, only to get criticized for having the slope of an access ramp be 1/4 degree off ADA standards and have a grievance filed by the union that park visitation had gone up, creating more work for the government employees.  I spent an evening having a beer with that manager, and you can bet they are never going to try to actually improve the customer experience again.  As another example, I went in to my government landlord last week and just blasted them for their lack of customer service focus, for the fact that they are blocking me from making improvements customers are begging for.  They yawned, gave me no response,  and handed me a notice that they were missing some of our water testing paperwork and please get it to them ASAP.

NCLB just gives government schools another government wammy to be managed and avoided.  The authors will probably get their wish, and huge bureaucracies will rise up to manage the numbers and reports without anything being done to really improve education.  The authors lament that the California state education department has "only" 1452 employees.  I have every confidence that this "problem" will soon get fixed by California, and the number will balloon up nicely, long before children see any better education.

A while back I wrote a plea to just let GM die.  I said:

A corporation has physical plant (like factories) and workers of
various skill levels who have productive potential.  These physical and
human assets are overlaid with what we generally shortcut as
"management" but which includes not just the actual humans currently
managing the company but the organization approach, the culture, the
management processes, its systems, the traditions, its contracts, its
unions, the intellectual property, etc. etc.  In fact, by calling all
this summed together "management", we falsely create the impression
that it can easily be changed out, by firing the overpaid bums and
getting new smarter guys.  This is not the case - Just ask Ross Perot.
You could fire the top 20 guys at GM and replace them all with the
consensus all-brilliant team and I still am not sure they could fix
it. 

All these management factors, from the managers themselves to
process to history to culture could better be called the corporate
DNA*.  And DNA is very hard to change.  Walmart may be freaking
brilliant at what they do, but demand that they change tomorrow to an
upscale retailer marketing fashion products to teenage girls, and I
don't think they would ever get there.  Its just too much change in the
DNA.  Yeah, you could hire some ex Merry-go-round** executives, but you
still have a culture aimed at big box low prices, a logistics system
and infrastructure aimed at doing same, absolutely no history or
knowledge of fashion, etc. etc.  I would bet you any amount of money I
could get to the GAP faster starting from scratch than starting from
Walmart.  For example, many folks (like me) greatly prefer Target over
Walmart because Target is a slightly nicer, more relaxing place to
shop.  And even this small difference may ultimately confound Walmart.
Even this very incremental need to add some aesthetics to their
experience may overtax their DNA....

Changing your DNA is tough.  It is sometimes possible, with the right
managers and a crisis mentality, to evolve DNA over a period of 20-30
years.  One could argue that GE did this, avoiding becoming an
old-industry dinosaur.  GM has had a 30 year window (dating from the
mid-seventies oil price rise and influx of imported cars) to make a
change, and it has not been enough.  GM's DNA was programmed to make
big, ugly (IMO) cars, and that is what it has continued to do.  If its
leaders were not able or willing to change its DNA over the last 30
years, no one, no matter how brilliant, is going to do it in the next
2-3.

I would say the exact same thing is true of public schools: Their DNA is senescent.  Most are the equivalent of alcoholics who keep falling off the wagon and keep asking for more chances.  At some point, you just have to give up.  At some point, it is easier to just start from scratch.  After 30 years of trying, Sears still can't change itself so there is Wal-Mart.  After 30 years of trying, GM still can't change itself so there is Toyota.  After 30 years of trying, United Airlines still can't change itself so there is Southwest.

The only difference in education is that the government has to date suppressed the emergence of Toyota and Wal-Mart and Southwest because, well, because it can.  I am sure that United Airlines would have liked to ban competition from Southwest, but it does not have the coercive power of government.  Fortunately, in most industries other than education, the public gets a choice of offerings, and companies that customers don't prefer tend to die.

It's time to give school choice a chance, and radically shift the incentives for public schools in a way that the government can't with bureaucracy-based programs like NCLB.  Some public schools will thrive, and many will die in favor of private options, but our kids will be far better off either way.  It's time to stop doubling down on failure.  It's time to stop giving the alcoholic one more chance.

Postscript:  One of the reasons that competition is important is in the very definition of "expertise."  An expert is someone who presumably has been succesful at a certain activity when others have been less so.  We call Herb Kelleher an expert on airlines and customer service because he designed a model that kicked everyone else's butt.  But would you have called him an expert in 1972, before Southwest took off?  Probably not.  He was just one of many voices with diverse, untested opinions of what would make a better airline.  What eventually made him an expert, and the others less so, is he went out and applied his ideas and they were succesful.

So the author's want to send more "expertise" to the schools.  OK, who are the experts?  Nearly every public school is using the same version of the same failed model.  Some succeed more than others, but these differences tend to be incremental rather than radical, like the difference between Sears and Montgomery Ward rather than between Sears and Wal-Mart (or even Amazon.com).  So how can you even know who the experts are within the same failed system, where no one is really allowed to go out and fully test their ideas in practice?  What happens, in reality, is that "experts" in education are the ones that can best enthrall academics and politicians and think tanks with grandiose or politically correct visions.  I would argue that as of this moment there are no experts in education in the US and we have no hope of identifying them until we let entrepreneurs go out and start testing various new models.

According to the WSJ($), you and I are going to take on the pension obligations of UAL:

A bankruptcy judge approved a
proposal from United Airlines parent UAL Corp. to transfer four
underfunded employee pension plans to the federal government, paving
the way for the largest pension default in U.S. corporate history.

The plans, which have a shortfall of $9.8 billion,
cover more than 120,000 United workers and retirees. United, the
nation's second-largest carrier in terms of traffic, wants to transfer
them to the federal Pension Benefit Guaranty Corp., or PBGC, which
would add to the already heavy strain on the agency from a spate of
pension defaults in recent years. Since accounting for United's
obligations last year, in anticipation it would assume them, the agency
has taken on obligations exceeding its assets by $23.3 billion  [ed note- the agency takes in only about $1 billion a year in premiums, so $23.3 billion in the hole is a very big number]....

The court's decision could have wide
repercussions in the airline industry, which is struggling with high
fuel costs, intense fare competition and overcapacity. Sidestepping its
pension liabilities will help UAL attract additional funding, while
giving it a huge cost advantage over many of its rivals, which are
saddled with underfunded defined-benefit retirement plans of their own.
That will put further pressure on those airlines to slash their costs
or in some cases seek bankruptcy protection in hopes of terminating
their own pension plans.

It is difficult for me to even start on how much this pisses me off.  These pensions are real obligations that UAL took on, and represent value provided in exchange for work that has already been done.  As outlined below, I am not big on the defined benefit pension model, but that does not change the fact that these companies are defaulting on a solemn obligation.  The temptation I guess is always great when finances get tight to defer obligations that are the farthest in the future, and so pension underfunding is one of the first things to occur.  There is no way management should get a pass for this, and I am flabbergasted that equity holders expect to retain anything out of the bankruptcy when employees have not been fully paid.

This being said, there is plenty of blame to go around, including for the union and the government.  The UAL unions should have been dropping the hammer on the company in the form of strikes or whatever at the first sign of under-funding.  Instead, they were more concerned about jacking up their salaries to the highest levels in the industry, ignoring the reality that airline finances by the late 90's were basically a balloon that if you pushed on it in one place, it popped out in another.  Unions allowed the underfunding to continue in large part lulled by the promise of the PBGC and taxpayers to make the pension funds whole if they continued to be underfunded.  This is the moral hazard that occurs in any kind of financial insurance like this, and the unions apparently were both right and wrong - we taxpayers will take on the obligations but their benefits will also get a haircut.

One of the lessons I thought was learned from the S&L bailouts of the 90's was that you can't provide such financial insurance without a parallel regulatory structure to make sure some kind of minimum fiduciary responsibility exists.  But, not learning a thing, the government has this pension guarantee program in place and exercises virtually no oversight over the funding or management of the insured pensions.

It is astounding to me that a large number of people still support defined benefit plans over defined contribution plans. What I don't honestly understand is why the rank and file still buy into this.  Defined contribution plans are much easier to monitor and audit and keep companies honest.  Once the money is in a vehicle such as a 401K, the money can't be taken away by the company or lost in a bankruptcy (unless the 401K is invested in the company's stock, which any adviser will tell you to never, ever do (see "Enron").  Now, I understand that there can be some tricky migration issues from one system to another, and companies use the transition as an excuse to cut back on their net contributions, but these are workable and negotiable issues .  My guess is that the support for defined benefit plans comes mainly from union leadership, since these plans give
them control of huge amounts of funds and thereby gives them extra
power (see Teamsters for the classic example, or more recently, the situation at Calpers).  I wrote more on this topic here.

The issues here are surprisingly similar to the Social Security debate, as discussed here.  Would you rather have the money in your own account, despite the fact you will then have to bear market risks, or would you rather the money remain in the hands of your company or your Congress.  In entirely parallel situations, money entrusted to UAL management and to Social Security has all been spent, with nothing now left to pay retirees. 

Update:  It just occured to me to ask - why don't frequent flyer mile holders ever have to take a haircut in an airline bankruptcy?  We frequent flyers are creditors too, holding a claim on the company in the form of our miles.  In fact, I would think my claim as a holder of miles is much much worse than other creditors.  For example, why should employees have their pensions cut before I get my miles account cut?  Heck, employees seem to have a much better claim than I do, especially since many of my miles were earned, like everyone else's, as marginally ethical kickbacks directly to me for influencing my employer's spending on air travel.  Despite this, it appears that pensions will be cut, and salaries will be cut, and bondholders will lose value, and stockholders will be diluted, but my miles will all still be good.

Update #2: Assymetrical information has a nice post along the same lines, pointing out an issue with corporate defined benefit pensions that I forgot to mention:  If you are 20 years old with a company, are you really willing to make a bet that your company will even exist in 60 years to pay off your pension?  Not to mention the portability issues, since few people remain with the same company to retirement.  I think I actually have a couple of defined benefit pension plans I am vested in from early in my career - one from Exxon, when I was about to quit to go back to school and was offered, due to poorly structured plan rules, the chance at early retirement instead.  I think I qualify for like $1.23 a month for life from that plan.

More also from Will Collier:

I don't mean to tread on Martini Boy's turf here, but the pensions
crisis among all of these old-line companies illustrates a great no-no
of long-term investing: lack of diversification. In the end, even
though they presumably didn't have much choice in the matter, all those
UAL employees who've been promised a defined-benefit pension are in the
same boat as the Enron and WorldCom employees who voluntarily put all
of their 401(k) money in their own company's stock. They bet the house
on one horse, and by they time old age caught up with the grizzled nag,
there was barely enough left of it to cart off to the glue factory

Kevin Drum also points out that these defined-benefit funds are easy to manipulate, since managers can play with the "expected returns" variable to change the necesary annual contribution. 

Recently, the government announced that it would take over the United Airlines pilots pensions in the government-funded Pension Benefit Guaranty Corp.  This move is irritating pilots, because their pensions get reduced, and it is annoying to me as a taxpayer, that I have to bail out a company that was too screwed-up to fully fund its pension obligations. 

This points up the biggest danger of government guarantees -- it causes companies to be more reckless.  Back in the 80's, banks and S&L's made insanely risky investments with bank deposits.  The people who should have been most interested in this problem - bank depositors - ignored it because they felt safe that the government had guaranteed their deposits.  In the same way, airlines and other ailing businesses with defined benefit pensions cut back on pension funding when times were bad, and the very group that should have been crying foul - the company unions - did not, because they again counted on a bail-out.

I put the blame squarely on the company's management, who made a commitment to employees and then failed to keep it, and now are using government pension gaurantees as a subsidy to close their cash flow gap.  However, it is interesting to look at the role of unions too.  For decades, unions have demanded defined benefit pensions (ones that promise a fixed amount per month at retirement) and have opposed defined-contribution pensions (ones where the company promised to contribute a fixed amount today into an investment fund).  I assume the main reason for this is that unions do not want workers to bear the market risks on investments.

Over time, though, defined benefit plans have, despite this opposition, gone the way of the dinosaur (at least in private companies - most government jobs still have them).  This is for a number of reasons:

• 401-K accounts now offer much of the same tax-deferral benefits for defined contribution programs that defined-benefit plans had
• Defined-benefit plans turn out to have market risk too.  One is inflation - benefits levels may be guaranteed, but unexpectedly high inflation can effectively reduce them, while defined contribution plans, if invested correctly, will likely produce returns to offset these inflation losses.  In addition, during go-go stock markets, holders of defined-benefit plans found out that they did not enjoy the benefits of higher investment returns - their employers pocketed them (by the way, may Americans are discovering the same about their Social Security benefits).
• As employees move around more, workers have found that defined benefit plans are not very portable, and tend to punish workers who do not stay for decades.  401-K plans are much more beneficial to workers who do not stay their whole career, or at least 20 years, in one place.
• As United pilots have found, defined benefit pension plans are hard to police by current employees- there are just too many variables that allow companies to argue that the pensions are OK.  On the other hand, defined contribution plans are very easy to police- one can check the amount of contribution each month against the amount promised.
• Finally, defined benefit plans rely on their company staying in business and fiscally sound for decades into the future.  This may have seemed a good bet at US Steel or United Airlines in 1950, but would anyone make that bet today?  For any company?

Marginal Revolution cites a James Surowiecki article on branding, arguing that increased information flow, particularly over the Internet, is reducing the power of brands.  This seems right to me.  Brands exist and command premiums for many reasons.  One role of brands is that they serve to reduce risk - without any other information about a product, many people would likely assume an electronics product from Sony to be more trustworthy than a no-name brand with the same features, and might be willing to pay a premium for the Sony product. However, with all the review information on the Internet, people may be more comfortable buying the off-brand, if it has good reviews, and saving the Sony premium.

Of course, brands serve some communication roles that are likely not threatened by the Internet.  For example, high end brands like Prada or Gucci have power because they allow the owner to communicate things about themselves to others.

I would argue that, even with Internet reviews, brands will continue to be powerful in the service sector.  In fact, with the growing complexity of some service offerings and the increasingly high standards of consumers, they may be more important.  Why?  Consistent product quality is much easier than consistent service quality.  A no-name product maker can get high quality product all over the world from one single factory -- all they have to do is to get that one location right.  This is much easier to do than with McDonalds, where there are thousands of locations, or even in our business, where we have hundreds of locations.  Service quality happens in real time, often in many dispersed locations miles away from supervision and the management staff. 

Also, in many cases, service failures are more critical and are harder to correct than product failures.  If my printer does not work, I get mad and box it up and return it for a new one.  But what happens if FedEx fails me on a critical shipment?  Or worse, what if United Airlines fails on me mid-flight? 

An interesting way to prove this is to go to a site like epinions.  Service reviews are generally much more variable than product reviews.  Compare Fedex, who's review is a mix of the lowest and highest scores, with this Apple Ipod, where reviews are much more consistent.  Even when products get a mix of low and high scores, often the low scores are driven by service and support and not the product itself.  In positioning their brand today, does Dell emphasize the product or their service around the product?