---

Congress passes a disaster relief package without the additions concerning government shutdowns and census taking that were included in the original bill. What changed their minds? Kwame Holman reports.

---

JIM LEHRER: The end of the great disaster relief standoff is first tonight. Kwame Holman begins with this update.

KWAME HOLMAN: Speaker Newt Gingrich went to the House floor this morning to report that Congress was back on track to pass disaster relief aid for flood victims.

REP. NEWT GINGRICH, Speaker of the House: Let me just say to all of my colleagues on both sides of the aisle that we are making progress; I have been deeply committed to getting the flood aid to the victims I visited in Minnesota and North Dakota. I know how important it is to get them the aid. I was very disappointed when the President vetoed the flood aid on Monday. We believe we are very close to having it worked out and hope in the next few hours to be able to announce and then move a supplemental appropriations bill to provide the flood aid.

KWAME HOLMAN: Five and a half billion dollars in disaster relief aid, plus nearly two billion dollars for military peacekeeping efforts were the main parts of a supplemental appropriations bill vetoed by the President Monday afternoon. He did so because the bill also included two controversial non-appropriations items: an automatic continuing resolution that would prevent a government shutdown this fall if another impasse develops over spending issues and a ban on sampling methods, a technique the Commerce Department wants to use, rather than a headcount, to conduct some of the year 2000 census.

MIKE McCURRY, White House Press Secretary: We just are waiting for the Republicans to come to their senses and present the President a good, clean bill that gets the funding where it needs to go and doesn't gum up the works with extraneous measures.

KWAME HOLMAN: And public opinion seemed to be on the President's side. A CNN-USA Today poll this week showed 55 percent of those surveyed blamed the Republicans for holding up the disaster aid, while only 25 percent blamed the President. Senate Minority Leader Tom Daschle from South Dakota--one of the states hardest hit by recent floods--had blocked all other Senate action this week until the disaster aid issue was resolved.

SEN. TOM DASCHLE, Minority Leader: I think that in the last 48 hours it's become abundantly clear that there is extraordinary desire on the part of people from all over the country to have Congress finally act and finish their work. And I'm hopeful that at long last today we can announce to the country that we've done just that.

KWAME HOLMAN: This afternoon Senate Majority Leader Trent Lott came to the floor to announce that an agreement, indeed, had been reached on a bill that provides all the money for disaster relief, peacekeeping, and other items included in the original bill.

SEN. TRENT LOTT, Majority Leader: I think it is important that we get the disaster funds through and the funding for the Department of Defense Bosnia activities, but this bill has grown like topsy. There's no need for it to be $8.6 billion. There's been a lot of add-ons on both sides of the capitol of both parties.

KWAME HOLMAN: Some Republicans in the House weren't happy either, because the new agreement also strips out the provision affecting a government shutdown to be voted on separately at a later date. And the original language in the bill to ban the use of sampling in the year 2000 census has been altered to say only that such a technique will be studied further and a report issued before it can be implemented.

REP. BOB LIVINGSTON, Chairman, Appropriations Committee: It was unfortunate that it had to be this sloppy and as ugly as it was, but now it's over.

KWAME HOLMAN: This afternoon the bill easily passed both Houses of Congress. The disaster relief bill now is being sped to the President's desk for his signature.

JIM LEHRER: Now, some political perspective on this conclusion from Congress watcher Norman Ornstein of the American Enterprise Institute. What was the force that brought this about, Norm?

NORMAN ORNSTEIN, American Enterprise Institute: Well, it's not unusual in Congress to take a must-pass piece of legislation and add on all kinds of extraneous things to try and gain leverage, hope that you can force the President to hold his nose and sign something that he has to sign.

JIM LEHRER: Because he wants the other things so badly.

NORMAN ORNSTEIN: He wants the other things so badly, and to believe as well that you can have the upper hand in public opinion and force action on something that would be obnoxious to the President. It's not unusual, but it's also almost certainly the case--and we look at examples in the past and this one--that when it comes to a public test, the President always wins. And in this case--

JIM LEHRER: Why? Why?

NORMAN ORNSTEIN: The President is a single force who has more credibility than dozens of voices inside Congress. He can--

JIM LEHRER: It doesn't matter who the President is.

NORMAN ORNSTEIN: It doesn't matter who the President is. He can grab the bully pulpit and in this case what we had as well was an example two years ago where Republicans had brought the government shutdown, had lost some credibility in issues of this sort, and now images daily of flood victims, innocent victims, waiting presumably for relief to come. Under those circumstances the President's voice saying give me a clean bill so we can help these people overwhelmed the Republicans saying he's the one holding things up.

JIM LEHRER: Yeah. And the irony, of course, was that one of the extraneous matters that caused the Democrats and the President, in particular, to be so upset was a ban on federal government shutdowns.

NORMAN ORNSTEIN: What an interesting twist. The Republicans would all acknowledge that in the 104th Congress their move to confrontation with the President was a disaster, shutting down the government, a political disaster for them. They wanted to put in a provision that would be a failsafe against future disasters but also designed to give them more political advantage and leverage in spending battles ahead. The President said no. In trying to avert that political disaster they walked right into another one of comparable proportions.

JIM LEHRER: And the--is there any question--well, there's always a question, but the polls had a lot to do with this. The public reaction was what drove this in the final analysis, is it not?

NORMAN ORNSTEIN: In a couple of ways, Jim. We know that there have been surveys done all over the place, national surveys, that show that about 55 percent of Americans believe that the President was on the right side here. Only about 25 percent believe that Republicans were. Republicans--and they went back home--remember, this was delayed over the Memorial Day recess as well--got hammered by their own constituents. Also, virtually all the editorial opinion in the affected states, the Dakotas and Minnesota, from Republican as well as Democratic newspapers, from politicians, Republicans as well, criticized the Republicans. They were on the wrong side, despite the fact that some of their allies, including some of the movement conservatives, were urging them to hold firm. They came to realize that there's a time to fold, and they folded.

JIM LEHRER: And, of course, one of the other--the other extraneous matters that the President, that he cited on his veto, was the--the census thing which part of the deal today was they agreed to study sampling, rather than to doing it, right?

NORMAN ORNSTEIN: There were two face-saving measures here. There's a long-term disaster relief and a short-term. They included the short-term, took out the long-term, so that they could talk, as one of the Republican members said, about saving money, and also they saved face by saying they would just study the census matter. But let's face it, the reality is they wanted two things, they got neither. They didn't get the census provision either.

JIM LEHRER: Well, speaking of the census, the question on the census has always been on the sampling versus the head count is--should there be head count of every person, or can a technique known as sampling also be used? We ask that now of Ronald Abler, executive director of the American Association of Geographers and a member of a National Research Council panel that studied this issue, and David Murray, research director of the Statistical Assessment Service--a study group on scientific issues. Mr. Abler, you support sampling. Why?

RONALD ABLER, Association of American Geographers: Let me say that the panel, of which I'm a member, supports sampling, and I don't disagree with that. And I would re-emphasize your point which was glossed over in the intro to the piece that it is not a question of head count or sampling. It is question of using sampling to supplement a head count when you get out into the last 10 years where the going gets rough in terms of trying to find people and enumerate people that just are difficult to find or don't want to be found at that point.

JIM LEHRER: It's not one or the other. I mis-spoke. I didn't mean--all right. Explain that situation, where sampling, in your opinion and the panel's opinion, when sampling is a valid thing to do.

RONALD ABLER: Sampling is a valid thing to do when you've made a good faith effort to enumerate or count as many people as you can in a given unit of territory that the census uses. Now, there are a variety of options as to when you would start sampling, would you do it when you've counted 80 percent of the people, or would you do it when you've counted 90 percent of the people and we're quite open-minded about that. There are arguments pro and con. What there is not an argument on--at least as far as this panel is concerned--is that sampling to supplement--and only to supplement--the best efforts that can be made at reasonable cost and in a reasonable time to do a complete head count is appropriate, necessary, and, indeed, the census will be flawed if they're not used.

JIM LEHRER: All right. Explain how sampling would work. Give us an example. And then use sampling. We got to one of these points, whether it's 80 or 90 percent, it doesn't matter, and then how sampling would work.

RONALD ABLER: Okay. One way that it would work would be say that you get to 80 percent with the normal methods of counting. Then at that point you might go and find--let's say there are 100 housing units in this territory that we're concerned with. We've got 80 of them. At that point we would find 10 more and basically double the characteristics of each one of those households--each of the ten we count--we wind up with a representation of the hundred which consists--hundred households in the tract--which consists of 80 we've actually counted, 10 more we've actually counted, but we inflate by 2 the last 10 households we count to get the hundred.

JIM LEHRER: Assuming--the assumption being that it's--you can fairly well expect it--there will be the same kind of make-up in these other houses?

RONALD ABLER: In small units of geography, yes, you can.

JIM LEHRER: Without having to knock on the door, or having to find them--

RONALD ABLER: Without having to knock on every door. I think the critical thing to keep in mind is that an attempt at a complete enumeration is what can be thought of as a sample. And it's a worse sample than what will occur if well verified, widely used statistical techniques are used to get at the people who for one reason or another either don't want to be counted, can't be counted, will get better results at lower costs. It's a combination. JIM LEHRER: Don't want to be counted or can't be counted and yet, must be counted.

RONALD ABLER: Yes.

JIM LEHRER: All right. Mr. Murray, you think that's a bad idea. Why?

DAVID MURRAY, Statistical Assessment Service: I think there are some problems there, and I think we've heard just one of them now, the notion that there might be a slippery slope. The 90 percent, 80 percent, 60 percent--at what point do we let sampling take over from the constitutional mandate to enumerate, to go out and actually touch and count each being out there? When we use the census, we're talking about the biggest enchilada of all government data gathering. It sets political power in apportionment and it sets political money in a form of federal programs that are deputed and mapped and allocated according to the census numbers. So the people who have reservations about this usually pick up three main themes: constitutionality, mistake, and mischief. And maybe it's the third one that's the most consequential.

JIM LEHRER: All right. But let's start with mistake. How is--how is the mistake element here? Doing sampling, compared with enumeration under the circumstances that Mr. Abler just laid out, how does that increase the chances of their not being mistakes?

DAVID MURRAY: We are transferring one kind of mistake, the undercount, the enumeration difficulties, which is probably about 1.6 percent of the population last time was very well--very likely not counted. For another kind of mistake, sample error. And in census tract tests that have been run the margin of error on some of these samples has turned out to be fourteen, twenty-five, thirty individuals out of every hundred in small census tract sampling. That's a real serious issue, the point being that an unvisited apartment building, a census taker no longer goes there. Suppose there's a hundred individuals there; there could be a hundred and twenty-five--

JIM LEHRER: Based on what the occupation is of similar apartments in the same area.

DAVID MURRAY: They take a sample somewhere. They project that sample back into it. It's an integrated count this time. The sampling results will be used to shape the total accounting of individuals and their distribution. The census does two things: It tells us how many beans there are in the jar, but it also distributes them. It distributes them geographically and demographically how many Asian Pacific islanders are there in Oakland. When you have this kind of margin of error, the prospect is there that you're actually going to be magnifying sample error to such a point that you don't know where to position people, along comes mischief. This is where the gerrymander guy tries to draw lines and ends up with proportional districts that may favor one part or the other.

JIM LEHRER: And is the sampling problem more acute when it--when it is down at the local level than it is at the national level.

DAVID MURRAY: Absolutely. When you have a local level tract sample margin of error grows as a consequence. When you finally get up to the larger accumulation of all districts, some of that sample problem disappears; however, try this analogy. What if your bank called you, and I trust my bank officer very much, but they're no longer going to give you a running account balance. They're going to estimate your balance over 10 years. They'll have a margin of error--extra hundred dollars here--minus a hundred dollars the next day but not to worry, Mr. Lehrer, your account will come out fine. Now, you've got to write a check today, and you want to know whether it balances. That's the problem--the apportionment at the local level.

JIM LEHRER: Is that the problem at the local level politically, Mr. Abler?

RONALD ABLER: No. Our problems in sampling the more specific you get in terms of the geography, the smaller the units that are involved. That said, I think we have to recognize that sampling is widely used in our society for all sorts of very important decisions. We estimate unemployment on the basis of samples. We estimate crop yields. We find these methods work exceedingly well. Most importantly, when we sample with a known technique, and in a known way we know what the error is. We know about that 1.6 percent that David just mentioned. When you try to do a complete count, you do not know what your error is.

JIM LEHRER: Why not? Why would you not?

RONALD ABLER: It is simply logistical.

JIM LEHRER: You don't know where the people are, you mean? Is it when you go to the door?

RONALD ABLER: You don't where they are. You don't know precisely how many people there are. We know how many--we know how many households there are very well. And the fundamental basis of the census is not wild guesses about how many people there are where. It is in a very exquisite and very detailed mapping of household use done in conjunction with the Postal Service and a variety of other sources by the Census Bureau. But if you think of an example of moving, I don't know how many people are in motion on a given day, moving from one residence to another, but it's the number of people in the United States removed, every year divided on the average by 365. That is going to be tens of thousands of people who are legitimately ambiguous about where they are and where they should be counted.

JIM LEHRER: So you figure that that sampling can account for that and what about that issue, Mr. Murray? I mean, people are just moving. It's a big country. A lot of people--

DAVID MURRAY: We're going to miss a few, and we're going to put them where we want them, and here's where the temptation begins; the temptation of a virtual America that after we go to an integrated sample, fold it back into the census. In some sense we've pulled up the anchor. We're not really sure who and where they are anymore, and the temptation is there, and both political parties have recognized this to let some computer modeler in the basement of the White House somewhere construct through a series of computer programs, a kind of deep blue count of how America is supposed to work. The assumptions of that modeler and the fudges and statistical manipulations that are potentially there is a very disturbing feature of the sampling aspect.

JIM LEHRER: And Norm, back to you on this. That is what he--what Mr. Murray just outlined was the fear that the Republicans had that caused them to put this--the rider on this bill in the first place, right?

NORMAN ORNSTEIN: One fear, Jim. The census interacts with politics in three ways. Mr. Murray alluded to them. Two of them involve districting. We have 435 House districts. They're supposed to have equal populations. We have to, one, allocate them among the states. And, of course, they're not all going to be equal. Every state gets at least one, but every ten years we measure the population of states and then using a formula allocate the seats. Will Pennsylvania get 33 seats, 32, 34? So that's one thing. The second thing is within the states, once they know how many seats they have, we create districts that are equal in population. And then there's a third issue which I'll get to in a minute about the allocation of money from federal grants. On the first two, however, these are big political questions, which states will have seats and where are the seats going to be districted means politics, and what Republicans are suggesting at one level is because most of this under-count is minorities. We're going to have population added to inner-city districts. We're going to have re-districting done in a way where some Republicans feared that they might lose 25 seats or more on the basis of this kind of sampling technique, then you moved to and that districts would go to places that are less hospitable to Republicans. And then you get to the third area which is--

JIM LEHRER: Now, the Democrats argue that those folks need to be counted, they're under counted, and they deserved to be counted; that it's the only way--sampling is the only way to get it.

NORMAN ORNSTEIN: That it's legitimate, and the only reasonable way to get to them now is with a sample because we failed to do it using the enumeration. The third area which is an important one, though, is the allocation of money. If you had education money going from the federal government out to the states, you're going to get more money going if there's more population, into districts. Districts want sampling. Less urban districts--and this tends to be more Republicans in this case--don't.

JIM LEHRER: So, in other words, what they did today in the House and Senate did not resolve this issue by a long shot?

NORMAN ORNSTEIN: No, it's not resolved. The irony here is that many Republicans are now saying in this era of fiscal austerity let's spend much more money, which is what it will require--

JIM LEHRER: Just one on one.

NORMAN ORNSTEIN: Where in the past, we've squeezed statistical programs.

JIM LEHRER: All right. Thank you all very much.

REGIONS TOPICS RECENT PROGRAMS ABOUT US FEEDBACK SUBSCRIPTIONS / FEEDS:

Support the kind of journalism done by the NewsHour...Become a member of your local PBS station.
PBS Online Privacy Policy Copyright ©1996- MacNeil/Lehrer Productions. All Rights Reserved.