Why do we have public schools?

Why do we have public schools? Is it to educate students or to maintain property values? Or perhaps it is to provide "day care" so parents could work. Or maybe its just a jobs program or a vehicle for community stabilization.
Education still remains the primary "acceptable" reason for public schools. Most debates usually find some way to tie things to education, even when seeking other goals. Any possible education 'side effect' will be given primary focus to make the goal more palatable.
Property values is one argument that is also fairly acceptable to make in the open. The reputation of a school and schools district can have a significant impact on property values. Reputation can also be a virtuous circle. Schools with a good reputation are more likely to attract families that care a lot about education. This in turn will produce high performing schools, which will further improve the reputation. In urban areas, reputation can make a significant difference in property values. In Sunnyvale, a house in the Cupertino district may sell for a a few hundred thousand dollars more than a similar house a block away in the Sunnyvale school district. This occurs even in cases where both schools' performance are similar. (And in spite of the fact that both school s feed in to the same high school district.) Cupertino simply has a better reputation and all of its schools exceed state standards. Sunnyvale's reputation is 'ok', with some schools exceeding state standards and some schools falling way behind. (The quality of the schools is more a reflection of the population demographics than education, with the more poorer schools primarily being in low-income areas.) The reputation of the Cupertino district makes it more desirable, driving up housing prices. The reputation of the 'good' schools in the Sunnyvale correlates with higher property values than in the areas with bad schools (though not as high as the Cupertino district.) This may serve as a 'feedback' loop to encourage the continued economic segregation.
Though how much do the schools themselves really impact property values? As long as a school in a high-value area is 'adequate', the reputation will continue to attract good students, and the test scores will continue to look good. A school would have to be exceptionally good or exceptionally bad to break out of the trend. Thus, education and property values diverge significantly. From a property values perspective, the main goal is to minimize risk. Bold education moves may make teachers unhappy, or harm scores. Running schools in uniform way, and throwing extra dollars at it to minimize murmuring will keep everyone happy.
Community stabilization seems to come out as an excuse for other arguments. People don't want their local school closed, in part because they worry about property values. Or teachers don't want vouchers or charter schools because they worry about jobs. "Neighborhood" schools are, alas, an endangered species. In urban areas, school choice, charter schools, magnet schools, racial balancing, and a host of other factors have lead to a reduction in true neighborhood schools. In suburban areas, it is often distance and consolidation that play the leads. Its cheaper to build one larger school. And if everyone will be driving their children to school, it is just as easy to go to a magnet school of choice. Past experiences in forced integration often drove the wealthy out of public schools (either by moving, or choosing private schools.) The results may have been slightly more integrated schools - but much less integrated districts. A key side-effect was the decline in the influence of schools as a community pillar.
And that brings us to the jobs program. The prime proponent of this argument is the teachers' union. Often the public assume that the union's goal is to improve education. The union is willing to play along with this belief for the political benefits. However, its fiduciary duty is to the union and its members. The union would be successful if nobody learned anything in schools - as long as the teachers had fulfilling secure careers. However, the union is unlikely to go to this extreme. Teaching is more satisfying when students are learning. In general teachers want what they think is best for their students' education. However, teaching remains for most, a job. Working conditions and job security are of key importance.
For the union, stable membership is also important. Though a college professor would be more than qualified to teach a high school AP course, in most states they would not be qualified to teach. They have to go through a teacher training program and pass a certification test. It probably doesn't do much to improve teaching ability, but it would make them more committed to public education (and thus to the union.) In California, even private school teachers have to go through the certification process. Most states also dictate minimum instructional hours and curriculum to be taught in schools. This makes moving from job to job more straightforward. A tenure and seniority process, on the other hand, encourages teachers to stay (while also limited the ability of districts to dismiss teachers.) The barriers to entry help to prevent too much competition in the field, while the relative standardization allow for many different job opportunities. Since schools are located just about everywhere, this is an ideal job program.
An ideal job program should not be confused with an ideal educational program. The need for public school teaching credentials is commonly accepted, though there are plenty of people opposed. As a policy, it probably does a good job keeping most total idiots out of public schools. On the negative side, it probably prevents a lot of really great teachers from becoming public school teachers. The seeming homogenization of teachers allows schools to be more like assembly lines. Once qualified, a teacher can be plugged in as a cog on the education line. The union makes sure the salaries do not dip too low, while the standardization prevents salaries from rising too high. This also reduces the application of creative energies. (Thus creativity in education is often shown by classroom decorations rather than innovative teaching plans.) This encourages greater equality, but reduces the possible areas for innovation (or even custom adaptation).
As an entity, it is unions' duty to fight against competition. Compulsory education laws with high dropout age help to keep the target audience intact. Mandatory curriculum and days of instruction help limit competing schools. Mandatory teacher credentials for private schools makes it even more difficult for other schools to appear. School vouchers must be attacked - though they may allow more money per student in the public schools - put it would also lead to elimination of public teaching jobs (and be an indictment of the quality of public education). And besides, if public education is that bad, you can probably find a rich foundation that will subsidize students - allowing them to leave the schools without any loss of revenue. (Washington D.C. schools have announced an attack on tenure, noting that public school tenure only benefits adults. Attacking a job program during an economic downturn may seem odd - but D.C. schools do have a really bad reputation. And the plan is also an "opt in" plan that gives huge raises - funded by a foundation - in exchange for giving up some tenure rights.) Finally, it is also important to grow the membership by encouraging policies such as class size reduction and public preschool.
Preschool brings us to the daycare argument. Public schools watch children during the day, allowing parents to work. The schools are a cheaper alternative to daycare, and provide learning to boot. After-school programs help to fill out the day. This argument is not as common as the others - probably due to conflicts. (If a goal is daycare, shouldn't school days match working days? Are students up to days that long? And who would work all that time? Would that help or hinder education?)
In the end, the multiple competing interests have produced a decidedly average public education system. Students get a good (but not great) education. Teachers have good (but not great) jobs. Property values are maintained at a consistent level, and parents get a partial day break from their children. Is this what we want in public education?