James Madison’s Case for Federal Education “Mandates”


If you’re looking for an education secretary, Mr. Obama . . .

(Guest post by Greg Forster)

In a letter to Wall Street Journal on Friday, Pete Hoekstra follows the well trod path of populist conservatives who demonize “federal” interference in education and demand that power be handed back to “local schools.” (To his credit, Hoekstra also mentions school choice.)

Populist right-wingers need to learn that teachers’ unions and their allies laugh all the way to the bank when conservatives demand “local control” and romanticize the “local school.” The unions have a hammerlock on local school politics. The further you go down the chain geographically, the more power they have. Nobody votes in school board elections except school employees and their families and friends. They hold the elections at inconvenient times precisely to produce this result. Thus, local communities typically have little practical control over their own school boards. The school boards consider the staff unions representing teachers and other school employees to be their primary constituents. When the demands of school staff interfere with the needs of students, the school boards favor the staff.

If you want to know why so many of our schools are run as jobs programs and don’t produce a decent education, “local control” is how it’s done.

There are worthy criticisms of NCLB. Interference with local control isn’t one of them.

The “mandates” of NCLB aren’t even mandates. They’re conditions for funding. The federal government gives states tons of (my) money to participate in NCLB. Doesn’t it – don’t I – have a right to ask states to provide transparent data reporting and measurement of outcomes in return? And if the states are getting a bad deal, they can stop taking the money.

Whenever I point this out, the critics respond that states can’t be expected to turn down federal money no matter what terms it’s offered on. Well, if so, then the problem here isn’t with the federal government, is it?

But there’s a larger philosophical issue here. People think that pure, unsullied federalism requires not only that the federal government excercise no coercive power over areas of state authority, but that it exercise no form of influence whatsoever.

This is false, and for my authority I appeal to the original “federalists”: the authors of the Federalist Papers.

Federalist #47 and #48 take up an argument advanced by “the more respectable adversaries to the Constitution” – namely, that the Constitution fails to create a true separation of powers among the legislative, executive, and judicial branches because it allows each brach to influence the others. The president can veto legislation, the Senate gets to vote on cabinet members, etc.

Madison points out that there can be none of those crucial “balances and checks” between the separate branches if they exercise no influence over one another. A proper separation of powers not only permits but requires that each branch have some substantial beachhead of influence within each of the other branches:

Unless these departments be so far connected and blended as to give to each a constitutional control over the others, the degree of separation which the maxim [that powers must be separated] requires as essential to a free government can never in practice be duly maintained . . . . It will not be denied that power is of an encroaching nature, and that it ought to be effectually restrained from passing the limits assigned to it . . . . Will it be sufficient to mark, with precision, the boundaries of these departments in the constitution of the government, and to trust to these parchment barriers against the encroaching spirit of power?

This passage is the indispensible context – the “backstory,” as they say in Hollywood – for understanding the much more famous Federalist #51.

In #47 and #48, Madison shows that “pure” separation of powers is insufficient. In #49 and #50 he considers and rejects Jefferson’s position that government encroachments can be resisted by frequent appeals to the people. In #51, he draws the conclusion: true separation of powers requires not that powers be totally separate but precisely that they must be separated and then mixed or blended:

As all these exterior provisions are found to be inadequate, the defect must be supplied by so contriving the interior structure of the government as that its several constituent parts may, by their mutual relations, be the means of keeping each other in their proper places . . . . The great security against a gradual concentration of the several powers in the same department consists in giving to those who administer each department the necessary constitutional means and personal motives to resist encroachments of the others.

Whereupon he launches into the famous passage about ambition counteracting ambition, etc.

The real argument of Federalist #51 is not that we need a separation of powers – that argument comes in #47 – but that we need a certain kind of separation of powers. Specifically, the kind that allows each branch to have some power over the other branches.

Now, obviously this is all in the context of the separation of legislative, executive, and judicial powers, not the division of powers among local, state, and federal governments. But it seems obvious to me that the same principle applies – it would be dysfunctional for each level of government to have no influence over the others.

Of course, most of the founders did not envision the federal government influencing states and localities by offering them money. But it was always one of the very few great weaknesses of the original Constitution that it failed to clearly deliniate the boundaries of local, state, and federal responsibilities, and to provide institutional mechanisms to shore them up. It seems to me that in the phenomenon of what might be called “conditional federal subsidies,” like NCLB, we have stumbled unwittingly into a not-too-bad mechanism for allowing the federal government to influence states without directly taking over their operations.

As I said, there are legitimate criticisms of NCLB. The 100% proficiency promise is absurd. My own position has always been that the really valuable contribution of NCLB has been the mandate for transparent data reporting and testing.

But to say that it violates federalism to have the federal government attach conditions when it offers subsidies strikes me as not only incorrect, but an open door to unchecked power for “local” constituencies like the unions.

26 Responses to James Madison’s Case for Federal Education “Mandates”

  1. matthewladner says:


    The low quality of “local democracy” in school districts is undeniable. States should require school elections of all sorts to occur on the general election day with uniform voting sites.

    The balance that you describe between Federal and State government, however, was fundamentally disrupted by the 16th and 17th Amendments creating an income tax and requiring the direct election of Senators. Routine fiscal blackmail and pushing of unfunded mandates on the states is clearly well outside of the scope of what the founders had in mind- a very unbalanced situation.

  2. Greg Forster says:

    I agree that “unfunded mandates” are bad – they represent the “coercive power” that I said in my post the federal government shouldn’t have over areas of state authority. But NCLB is hardly unfunded! That was more or less my point.

    Which brings me to your use of the phrase “fiscal blackmail.” I’m not sure what lies behind that. The federal government says to states, in effect: “I’m willing to give you $X if you do Y. If you’re willing to commit to do Y, I’ll write you a check for $X. If not, no hard feelings.” How is that “blackmail”?

    Of course you’re right that states can and should reform school board elections. The fact that they haven’t done so is all the more reason not to romanticize state power (as Hoekstra does) any more than local power.

    As for the 16th and 17th amendments, I agree that direct election of senators is bad and I’m at least open to the argument that the income tax is as well, but those are separate conversations from the merits of NCLB and the general subject of conditional federal grants.

  3. matthewladner says:

    I agree with you that NCLB is not an unfunded mandate. I also agree with you that we ought not to romanticize the quality of state and local democracy regarding education policy. We also however should not underestimate the ability of the feds to make a hash out of education policy. The unions can just as easily exercise federal influence as state, for example.

    Fiscal blackmail comes when the feds put conditions on pre-existing funding streams to boss the states around. Lower your speed limit to 55, or lose your federal highway money. If you want medicaid money, you have to follow our 800 page rulebook, etc.

    My point is only that the founders never intended to have a federal government that was routinely bossing the states around. They gave specific powers to Congress (albeit with some flexibility) but then included the 10th amendment to specifically say that powers not given to the Congress were reserved to the states, or to the people.

    I think if James Madison came back today, he would be horrified by the sorry state of local and state democracy regarding education policy, but I doubt any of the founders would feel comfortable about an attempt to manage education policy by the federal government.

  4. Greg Forster says:

    I don’t think it’s true that the unions can exercise power at the federal level “just as easily” as at the local level. In the earlier post that I linked to above – it’s here – I argued that union power is greatest at the local level and weakest at the federal level; their power gets larger the further down the chain you go. I wrote then: “As Hamilton, Madison and Jay (the “Three Founderos”) observed in the Federalist Papers, selfish interests will always find it easier to extract goodies from the public fisc in a whole bunch of little local places than in one big place. While centralization does provide one-stop shopping, it also creates more intense scrutiny and greater opporutnities for opposition.”

    It’s true that the founders never envisioned the current state of things – I said the same above. But my argument is that the founders never adequately addressed the structure of federalism at all. If the founders didn’t provide an adequate vision, as I think is the case, then we were always bound to end up somewhere that they didn’t envision. And the destination that we have in fact blundered into is not actually all that bad.

    Just for the sake of clarity, is it your position that conditional federal mandates are “fiscal blackmail” if the funding stream existed first and then the conditions were added later, but not if the funding and the conditions appear at the same time?

  5. Patrick says:

    I wonder if the federal government would even realize if the state was complying with the mandate or not.

  6. Greg Forster says:

    Depends on what kind of non-compliance you’re talking about. They’re better at enforcing some of the mandates than others.

  7. matthewladner says:

    Personally I think that if the education unions have less influence in Washington than in the states, it is only because they haven’t bothered as much with the feds. If we invested education authority with the United Nations, the unions would quickly have East 44th through 49th streets in Manhattan covered with giant lobbyist temples.

    I don’t have any illusions that anyone is going to overturn the 16th and 17th amendments or that the Courts will wake up one day and discover the text to the 10th amendment. I do however think that things have gone far out of balance, and that we should find ways for correcting that imbalance.

  8. Greg says:
    “Which brings me to your use of the phrase “fiscal blackmail.” I’m not sure what lies behind that. The federal government says to states, in effect: “I’m willing to give you $X if you do Y. If you’re willing to commit to do Y, I’ll write you a check for $X. If not, no hard feelings.” How is that “blackmail”? ”

    My answer:
    I think it’s blackmail because the check consists of money they took from you (or people in your state). If you (as a state) decide not to do anything for which the feds will cut a check, then you are being taxed and having that money go to people in other states. The pressure on you to do what the feds are suggesting is the pressure not to pay taxes and receive none of the revenues.

  9. Brian says:

    Look, my level of comfort with the fiscal blackmail that occurs usually depends on the issue. But everyone should acknowledge it IS blackmail. That statement with the Xs and Ys from above leaves out one important detail–it should read:

    “First, I am going to take X from you, and I’m willing to give you $X back if you do Y. If you’re willing to commit to do Y, I’ll give you your $X back. Otherwise I will keep your X”

  10. Greg Forster says:

    But the point that “they took the money from me in the first place” cuts both ways.

    The federal government is going to give states big money for schools regardless of whether we attach any conditions to it. So I’m going to lose that money anyway. The question is whether we’re just going to pour the money down the rathole and get nothing, or demand at least some kind of minimal accountability – even if only in the form of transparent data collection and testing.

    Since that money was originally my money, I’m all the more forward to say that as long as they’re going to take it from me and give it away, I’d prefer they got something for it.

    If the conditions attached to the funding weren’t related to the activity the funding was going to, that might change the argument. But if the conditions attached to the funding are clearly related to ensuring that the money is well used, I don’t see the basis of the complaint – unless you really think that as long as the feds are being stupid by redistributing money, they should go all the way and distribute it without trying to direct the money to good and effective uses.

  11. Greg Forster says:

    Neal McCluskey critiques this post, but I think he has misunderstood my argument.

    Contra Neal, my argument is not about what the Constitution permits as a matter of law, but 1) whether it’s desirable or undesirable for the federal government to exercise some influence in areas of state and local authority, independent of which means of influence are or are not permitted under the Constitution as a matter of law, and 2) whether what I call “conditional federal subsidies” perform this function in a desirable way. I cite Madison in support of my argument that it’s desirable for different levels of government to influence one another’s areas of authority. My point is philosophical and I cite Madison as a philosopher, not as an expositor of the Constitution’s textual meaning.

    The question of what the Constitution permits or doesn’t permit is not raised here – if it were, I’d have brought a different set of arguments. And I’d have cited the text of the Constitution, which is legally binding, rather than the Federalist Papers, which is not – Scalia is right that we should look at the “original public meaning” of the text of the law itself rather than the “original intent” of those who wrote it.

    To the extent that Neal’s concerns about the boundaries of what the federal government is or is not legally permitted to do appear in my post at all, I actually criticize the Constitution for providing an inadequate articulation of the boundaries of federalism and inadequate institutional arrangements for preserving those boundaries. So naturally it’s not a high priority for me, as it is for Neal, to suss out just exactly what the Constitution does or does not permit.

  12. With all the New Talkin’ I’m up to — and great moderating, by the way, Jay! — I have to make this brief. Let me just hit a point about unions and federal control. Of course, right now the unions focus more on state and local-level power because that is still where the bulk of education authority resides. But as authority moves on up the ladder so too will union influence. Indeed, the unions have long been trying to pull it up: witness the U.S. Department of Education, a gift to the NEA! And the protections of dispersing power about which the “Founderos” talked really no longer applies to the federal level because they relied largely on geographic dispersion. With airplanes, Blackberries and cell phones, that distance has been conquered and factions can easily collude. And if they conquer in Washington, there will be nowhere to hide.

    One last thing: When the feds take your money they don’t just return it with good stuff attached. They return what’s left after buraucratic processing and with lots of annoying strings attached. Oh, and they give some to the Historic Whaling and Trading Partners while they’re at it because, well, those folks need your cash, too.

  13. Thanks, Neal. I’m enjoying being the moderator over at NewTalk in part because I enjoy having something like the word “moderate” close to my name. It’s quite novel and refreshing!

    I made an argument very similar to Greg’s in the New Talk discussion yesterday to press Neal. I’m not entirely convinced, but it seems plausible to me that Greg is right that organized interests have much more power in state and local education arenas because hardly anyone besides school employees pays attention to or votes in local school elections. The potential advantage of having these issues decided at the national level is that all sides can more easily monitor developments and organize to insert their voice into the debate.

    Neal is right that cell phones, etc… make organizing at the national level easier and could allow coalitions to form more easily, but it also allows groups on all sides to monitor and act more easily.

  14. Greg Forster says:

    I think we need to distinguish between the economic argument against federal involvement and the federalist argument against federal involvement. Certainly the transfer of funds through federal hands involves a dead-weight loss of economic value, both because the feds take a large cut and because economic resources are distributed less rationally. If an end to federal involvement in education were a live option, the economic argument would be a good argument for ending it. What would not be a good argument, in my opinion, is the idea that the federal government has no business influencing areas like education because they’re reserved for the states alone. Lawmaking is reserved for Congress alone, but the president is still supposed to influence the process.

    The argument that selfish factions can extract goodies more easily at the local level does not primarily rely on geographic distance that technology can overcome, but on the lack of public attention. When unions take money from the federal budget, that will be noticed by at least some powerful people who will want to stop it. When unions take money from the budget of the Blatherboro School District, how many powerful people notice, other than those involved in the taking? If you have your tentacles in every local town in the country, it’s a lot easier to take $1,000 from each of a thousand towns than to take $1 million from the federal budget – no matter how many Blackberries people have.

    The creation of the USDOE was not an attempt to move authority over education up to the federal level but to turn on a new spigot of money. It was the anti-union reformers who decided that the flow of federal money created an opportunity to use federal power for reform. That’s what NCLB was all about. It was NCLB that represented an attempt to introduce federal power into education. All the unions wanted was money. They thought their friends would be in charge in Washington and they would therefore be able to get the money without ever paying a price in influence – more fool they.

  15. matthewladner says:


    Rod Paige made the statement at the begining of the Dubya administration that the federal government had spent billions and billions on Title I and had next to nothing to show for it. He of course made that statement in pursuit of NCLB, with the argument being NCLB would put a stop to that.

    But the argument cuts both ways. Now we’ve spent many more billions, and still have next to nothing to show for it.

    I would describe this idea that somehow Congress will make better education policy than the states the Mount Olympus Myth. The record seems to indicate that much more often than not the feds make a hash of things. The feds have been in K-12 education for 50 years now. Where’s the Beef?

  16. Greg Forster says:

    We have gotten tremendous value out of NCLB. Just not from the money we spent.

    See here for a brief review of the value NCLB has produced.

    If getting the feds out of education were a live possibility, this would be a much more complicated issue. Since it isn’t, the issue is simple: are we just going to throw money down the rathole, or are we going to throw money down the rathole while at least getting improvements in transparent data collection and testing?

    I’m sure the badness of federal involvement in education outweighs the good. But the question before the house seems to me to be whether we’re going to get both the bad and the good, or just the bad.

  17. Let me start by making clear that I’m not a defender of local public-schooling control. That said, I am not at all convinced that there is more transparency or opposition to special interests in Washington than local districts. For one thing, much more decision-making at this point occurs at levels above localities–especially states–than at the local level, and it does appear that people pay attention to state policy (to the extent that they pay attention to education policymaking at all). It is also not uncommon to have vocal and strong opposition to education bond referenda at local levels, which could never occur at the federal level because education gets its appropriations at the same time as heaps of other areas. And let’s not kid ourselves–in Washington there are no organized interest groups focusing only on education working for parents or regular taxpayers. The blob, in contrast, is all over the place. And transparency? How many people, do you think, have ever read NCLB in its entirety? Even better, how many do you think have read the regulations? And to shrink the influence even further, how many representatives of regular parents and taxpayers do you think have particpated in negotiated rulemaking? Finally, I think that Greg is right that the NEA pushed for the Department of Education first to get the money, but they also wanted a high-level agency they could control to push such things as requiring teacher input on local decision-making and force unionization.

    As for NCLB, I don’t think the benefits of, essentially, more data, have outweighed either the additional dollar costs or the continued centralization of power. We have not seen gains that can be attributed to NCLB on NAEP (indeed, it seems likely that gains were being made faster before NCLB), there are clear examples of states lowering standards as a direct result of NCLB (Michigan comes immediately to mind), and for the most part, the law has done nothing to encourage states to raise standards. Indeed, if anything, NCLB may have made finding the truth even more of a trip through a house of mirrors for parents. They have to know what has changed on standards, tests, and cut scores in order to know if their kids are really making progress, and NCLB essentially assures them that if their kids are deemed proficient, everything is OK.

    Of course, the same things could happen at the state level, so all that said, I think top-down control at any level is bad to just slightly different degrees. Which is why, ultimately, funding must run through consumers who bring it to autonomous schools.

  18. Greg Forster says:

    I don’t think it’s true that “much more” decisionmaking takes place at the state rather than local level. I don’t even think that “more” decisionmaking takes place there. I argue this point in that post that I’ve linked to above. Who negotiates the teacher contracts, where the factory-scale salary and tenure systems cause so many of the dysfunctions in the system? Yes, there are some states where the tenure system is imposed at the state level, but not everywhere, and at any rate the factory-scale salary system is more important and is local everywhere. Who does the actual hiring and (in theory) firing of the actual teachers, administrators, and staff? Who has primary responsibility for making sure principals and superintendents are running good schools? Etc. The amount of decisonmaking that takes place at the state level may not quite pale to insignificance next to the decisionmaking at the local level. But on the question of where more decisionmaking takes place, and especially where more of the decisions that cause problems take place, I don’t think there’s much doubt.

    You argue that the benefits of NCLB don’t outweigh the costs. Once again, I reiterate my point that we’re going to have the costs anyway. The only real question is whether we’re going to have the costs and the benefits or just the costs.

    I’m not sure I agree that the recent improvement in elementary-level scores on NAEP can’t be attributed to NCLB. But it’s a difficult question and I don’t think it has been well studied yet. The good news is, thanks to NCLB there are tons of data out there and I expect good studies will emerge over time.

    To your point about top-down control, as long as we have a government monopoly we’re going to have top-down control and all its delightful side effects. The only difference I can see is whether that “top” consists exclusively of the teachers’ unions and their lackies on school boards, or whether it will also include other influences – influences which may not be all good, but will certainly be less baneful than the unions.

    Thankfully, we can at least agree that there will be no real end to the dysfunctions until we get an education market!

  19. It is accurate that local districts have a fair amount of power over teacher salaries, hiring and firing principals and superintendents, and making ths buses run on time. But states decide on right-to-work status. States increasingly control education funding. States dertemine licensure requirements (though now with federal guidance). States often adopt textbooks. States set many standards (though again, under federal guidance). States create accountability regimes (of course, with federal guidance). And lets not forget, “local” has come to mean something much different in the last several decades. In 1939 we had about 120,000 school districts. Today, we’ve got about 14,000. Centralization, not greater localization, has long been the name of the game.

    As far as the costs and data go, it is wrong to assume we’d always have the costs. Yes, the feds spent a lot on K-12 before NCLB, but they spend much more now, and allowing them to do so was a political necessity to get NCLB passed (and I think we can expect the next president to spend even more). And I don’t think the data is much better now than if the feds had simply said that if you want federal money, you must particpate in NAEP. Yes, we have state tests, but they have largely been a moving target since NCLB was passed, making it hard to get consistent longitudinal data for many states. And for the people who are supposed to ultimately be using the data–parents–the difference between scores and reality can be very can be very hard to know.

  20. Greg Forster says:

    It’s true that spending went up significantly with NCLB. But the real cost of NCLB is not all that additional spending, but only the portion of additional spending that wouldn’t have happened anyway without NCLB. The whole reason NCLB happened in 2001 was becasue the stars were aligned at that moment for a big increase in federal education spending. Bush didn’t just want to impose a new federal accountability system on schools as a condition of federal spending growth. He wanted federal spending on education to grow independent of the accountability stuff. The money was not just the price he paid to get the accountability that was part of his “compassionate conservative” agenda. The money was also part of the “compassionate conservative” agenda.

    Just how much of the additional spending in 2001 is really attributable to NCLB we’ll never know, of course. But I have a high estimate of the benefits of the transparency and data produced by NCLB so I’m inclined to think we probably got a good deal compared to what we would have gotten if we had just gotten spending growth.

  21. I’m not so sure we would have gotten the same growth in the absence of NCLB. For Bush it was much more about “the soft bigotry of low expectations” than spending more. But to get the accountability piece done, he had to agree to bigger spending.

    Regardless, as you say, we don’t know what we woud have gotten without NCLB. And the important thing, of course, is that we agree on the primacy of school choice.

  22. matthewladner says:


    Take a look at the Proficiency Illusion study by Fordham. It really is sobering-much of the “data” that we have is highly dubious, and much of what we have we would have had anyway without NCLB. Granted, the state of school transparency is better now than it used to be and some of that can be attributed to NCLB. Both the Bushie 2014 utopians and the allegedly “less powerful on Mount Olympus” unions are actively putting these limited gains at risk.

    The reasonable assumption to make, in my view, is that the whatever progress we see in coming years will be driven by state policymakers regardless of what happens in DC. DC might hinder them somewhat, and they are not likely to help much either.

    In the end, the 8 percent of K-12 spending main result is paperwork, and that won’t change. The feds apparently cannot leverage their 8 percent into summoning the moral courage required to do real education reform at the state level.

  23. Greg Forster says:

    What the Fordham report shows is that the interpretations of the data are often dubious. States will set their cutoffs for achievement standards in ways that don’t make sense. But the underlying data are sound.

    In my study on Florida vouchers earlier this year, the only reason I was able to control for LEP status is because the LEP data were available in NCLB reports. They weren’t available anywhere else.

  24. matthewladner says:


    Whether or not the underlying data is sound (by which I’m guessing you mean usuable by researchers (?)) is secondary to whether parents are getting reliable information about whether their children are actually grade level proficient or not.

    Most state tests are too easy to do that, and then on top of that, many states lack internally consistent standards. While I think matters have improved from the days when real estate agents were the only source for school data, it’s not as though we’ve entered into a transparency nirvana.

  25. […] at Jay P. Greene’s blog, Greg Forster takes issue with those who say that the Constitution does not permit federal excursions like No Child Left […]

  26. Mindy says:

    I just could not go away your website before suggesting that I really enjoyed the
    standard info a person provide for your visitors? Is going to be again regularly in order to check out new posts

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: