you go give an audience, give a speech to 100,000 people? of course you would. well, that's what you're doing with c-span, with special orders every afternoon. so c-span became, i mean, he quickly becomes a cult political leader, and he's getting, you know, 700 letters a week from people around the country to this back bench, you know, junior member from georgia who's a member of a minority party. he's already achieving a national following. >> watch "after words" sunday night at 9 p.m. eastern on c-span2's booktv. >> commerce secretary wilbur ross testifies now about efforts to modernize census bureau operations ahead of the 2020 census including information technology investments. also the bureau's efforts to reach out to traditionally underrepresented communities and containing costs.

>> the committee will come to order. without objection, chair's authorized to declare a recess at any time. secretary ross is with us, he'll be joining us momentarily. i'll recognize myself for an opening statement and then my friend from maryland. the census is constitutionally required. it is what insures that just as every vote should be counted, every person should be counted. this is a conditioned precedent for accurate redistricting and accurate reapportionment. the census is not only important for purposes of constituting our government, it's also important for apportioning the resources and services of our government. so the census is vitally important for a number of reasons. it is also eminently predictable. there are a few things that can be foreseen quite like the census. it happens every ten years whether you're ready or not. so it's incredibly important beginning with the fundamental truth of counting every one of our fellow citizens, and we know well in advance when it's

coming. so we can't complain that we were caught off guard, and we can't complain that we didn't see it. so we have a couple more years before we actually count everyone, but now is the time to make sure that we are ready to count everyone with accuracy, with security and with professionalism. we're all concerned with cost. our fellow citizens are hearing about cuts in a number of areas, and there's concern about the deficit and the debt, so cost is important. but it's also very important that we get this done right, accurately, professionally with security and in a fashion worthy of the world's great democracy. what would be even worse than unexpectedly high costs is unexpectedly high costs coupled with less than full success. so we want to make sure the census is a success. we don't want to appropriate money on this committee, but we do help make sure the right questions are asked in a timely fashion. so for secretary ross, i appreciate his time and i appreciate the time, frankly,

that he's given our committee before today as well as his appearance today. we live in a countries that is increasingly -- country that is increasingly skeptical of whether or not we can get things done right, in a timely fashion and in a credible, responsible way. i think our country as a whole, irrespective of the region in which you may live or your political ideation, would appreciate a good news story. a story of how a government entity was prepared, efficient and exceeded their expectations. so 2020 may seem like a long ways off, but it's coming. and when it comes, we're going to have to have counted 303 million plus people accurately. so i think time is of the essence even today. so with that, i would welcome all of our witnesses, especially secretary ross, and would yield the remainder of my time to my friend from north carolina, the chairman of the subcommittee, mr. meadows. >> thank you, mr. chairman. thank you for your leadership on this particular issue. and as we go into this hearing

today, i think it's critically important that we all realize that while 2020 is a good ways off, we have already missed critical dates, critical testing and critical implementation that should have already been done according to our own schedule that we have had in previous hearings. of the i.t. systems that are out there, there are some 43 i.t. systems that need to be implemented. and yet today only four have actually been fully implemented. another 31 are only partially done, and 18 have not been delivered in any form or fashion. and yet we continue to have hearings over and over and over again about the critical nature of making sure that we not only get these systems implemented, but tested and that we do

end-to-end testing on a basis to make sure that cybersecurity, other fumbles do not happen. and yet here we are today on another hearing, and i am concerned that the focus will be on the $3 billion in terms of a budget shortfall and not on the other critical mission steps that have to be addressed. and so let's today focus not on the $3 billion, but on the management issues that must be done. and in doing so, we're talking a whole lot about disaster relief and every other area. today is a day that we need to start talking about disaster relief for what will be a disaster if we don't get on this today. and so, mr. secretary, thank you for coming. for our second panel of witnesses, thank you for coming.

i've been informed by some of your staff, mr. secretary, that you are on this and that you plan to make sure this is not just a budget tear component -- budgetary component, that it is a component that actually looks at the management, from curement and otherwise. and i was pleasantly assured of that this morning. but we look forward to hearing more from you. i thank you, mr. chairman, for your leadership. and as we get this right, i look forward for this to be a good story in the months and years to come. i yield back. >> gentleman from north carolina yields back. the gentleman from maryland is recognized. >> thank you very much, mr. chairman. if there's one key point we can get across to those watching today's hearing, i believe it is simply this: the census bureau is dangerously underfunded and has been for years. and unless we do something about it right now -- not yesterday,

not today, but right this moment -- this massive deficiency could imperil the fairness and accuracy of the census itself. americans expect us to be able to do things right, and i think the chairman would agree with me, i think he just said it. we want to be effective and efficient in whatever we do. mr. meadows is right. we've got to -- it's not just a thing of money, it's a thing of operation and is making sure that things are done properly. so this is not a partisan observation. both progressives and conservatives agree that current budget projections are way, way, way, way, way too low. the president of the r street institute wrote an op-ed just this tuesday warning that the

census is, quote, understaffed and underfunded, end of quote. and that, quote, america is in serious danger of the census going wrong. end of quote. ladies and gentlemen, we cannot afford to allow that to happen. this is the united states of america, the exceptional nation. if we cannot count the 300-plus million people who are in our country, i don't know how exceptional we'll be determined to be. last month the conservative american enterprise institute joined with the center on budget and policy priorities in warning congress that current level of flat or near flat funding, quote, is up precedented and would -- unprecedented and would significantly undercut efforts to conduct an accurate survey,

end of quote. they warned that, quote, shortchanging the census is penny wide and pound foolish. that it, quote, jeopardizes the bureau's ability to implement cost effective new technologies, end of quote. and it, quote, could backfire, costing taxpayers more in the long run. end of quote. i say that we are better than that. i'd like to show a few charts that illustrate this urgent problem. it won't show up on screen but, members, it will show up on your screens at your desks. first, as we all know, funding for the census bureau certainly increases dramatically in years 8-10 of each decade to accommodate the ramp-up in activities for the census in year 10. the first chart shows funding for the last four decades in inflation-adjusted dollars. for the 1990 census, the bureau

increased from $596 million to $3 billion over the last four years. for the 2000 census, it increased from $535 million to $6.9 billion over this period. for the 2010 census, it increased from $1.1 billion to $8.4 billion over this period. but what are we doing now? we are basically flatlining. the trump administration did not request a significant increase for 2018. you can see the same trends when you look at the budget from a percentage basis which is what this next chart shows. the 1990s census had a massive increase in the last four years of the cycle. as did the 2000 census. and the 2010 census, but here is

where we are now. the administration is not on track to match past increases for 020. 2020 . if we use these historical trends to project forward, total seven cuts bureau funding -- census bureau funding would be increasing over the next several years. but that is not what is happening. as this last chart shows, there's a shortfall of about $760 million when you compare what the administration is requesting to past trends. i say that we are better than that. now, congress shares part of the blame, no doubt about it. in 2012 congress insisted that the census bureau spend less on the 2020 census than it spent on the 2010 census. congress refused to acknowledge or budget for two of the most senate and predictable drivers

of cost increases that occur between each di seven y'all census, inflation and population growth. aggravating this problem even more, congress appropriated even less than the census bureau requested in every single year since 2012. now, some people will argue that the census bureau was supposed to save billions of dollars this time around by using new i.t. innovations to reduce the number of workers it needs to hire and increase the number of people who submit their forms online. the problem is that when you starve the census bureau year after year after year, it cannot make the investments needed to implement these innovations. ..

explain to the washington post just this week, had we been funded to do everything we asked for then we would be much farther ahead. he continued, quote, at this point they have to go back and do some of it the old way with her and pencil. end of quote. with paper and pencil. this is the exceptional nation, the one that over and over and over again. i want to thank you for being here today and briefing members of the committee. i understand you have limitations on your time but we beg you -- the chairman will emphasize this.

we have members that are very interested in this subject. unless your meeting with the president i hope you can spare a few moments because we want to have members ask and as i close i know you will get into this during your testimony but i appreciate your own team's review you recommend the administration increase its request for 2018 x 100 $87 million and thank you for doing that because so often we have secretaries come in and don't say what they need to get people upset. this is too important. i commend you for taking the step and you understand the.of this issue as a former enumerator yourself. even this request is too low. too low. especially given the number of

tests have been canceled and extremely low number of hires at the moment. i know you are constrained, the office of management and budget rejected your request for $50 million for contingencies that are certain to occur. nevertheless we have a responsibility under the constitution of the united states to ensure the census is funded adequately and for census bureau has resources to conduct a fair and accurate census. this should matter to each and every one of us. why we certainly discuss minority and immigrant populations, low income and rural communities are also at risk of being missed particularly in a digital

census. i joined my colleague representative maloney who chairs the conventional census caucus, to fund the bureau at 1.93 $5 billion this year. an amount that reflects budgets from the bureau's fiscal year 2017 budget and an extra $135 million for increases in funds to the seek apps program. we recognize the gravity of the situation we are facing. you have taken this matter very seriously. you said earlier we had a very important meeting with the secretary. that helped lay the foundations of this hearing and i look forward to your testimony, thank you for your indulgence. >> we want to welcome you,

mister secretary, pursuant -- we will administer the oath. raise your right hand. do you solemnly affirm the testimony are about to give is the truth, the whole truth and nothing but the truth so help you god? may the record reflect the witness answered in the affirmative. take your seats. your entire opening statement will be made part of the record. you have five minutes to summarize that. as mister cummings said we appreciate that there are other demands on your time and we want to be good stewards on your time. this is one of the unusual entities that is squarely in the jurisdiction of this committee and lots of members have questions. what i ask of my colleagues on both sides is to exercise

self-restraint of getting your questions done in five minutes. i am notoriously slow on the gavel, hard to unlock the mysteries of the world in five minutes but i ask my colleagues to self-discipline, get it done within five minutes so everyone has a chance to interact with you and to be a good steward of your time. i invite you to turn on the microphone and you are recognized for your opening statement. >> thank you very much. ranking member cummings, members of the committee, thank you for your continued support, your discerning oversight, thank you also for allowing the opening statements to be read while i was reviewing the newly developed material to comment on it myself. and official 20/20 census that provides a full, fair and accurate count has been one of

my highest priorities since being confirmed in february. i myself was an enumerator as i worked my way through business school so i have a strong appreciation for the responsibilities and unique challenges of counting everyone in the united states once every ten years. any operations that must hire and manage half 1 million temporary employees with complex and new technological systems and rigid completion dates is bound to be a difficult undertaking for even the most experienced manager. it has been clear to me from the beginning of my tenure at census would be one of the most challenging aspects of the commerce portfolio. i agree with members of this

committee, the census is the bedrock upon which we construct our system of democratic representation. it provides redistricting, the distribution of hundreds of billions of dollars of federal funding. accordingly the first meetings as newly confirmed secretary of commerce with the commerce staff. in those meetings i sought to identify and address the key issues faced by the decennial census. my early concerns were heightened when only two months into my tenure for census bureau suddenly announced a 40% cost overrun in one component, namely the census enterprise

data correction and processing program. a critical part of the technology infrastructure for the 2020 census. in terms of the broader decennial census the prior administration lifecycle cost estimate, october 2015, was $12.5 billion. when testifying before congress in june i did not accept that figure. instead, i vowed to return to congress after a thorough review with a fetid 2020 census lifecycle cost that i could support. the product of that review is what brings me here today. as promised we assembled a team of experts to conduct an independent review of the estimates. they have come back with

numbers as we continue our preparation for 2020. team is comprised of financial management experts from the department of commerce, the office of management and budget, former census employees, two former technology executives with experience rolling out complex systems as well as other experts with extensive private-sector experience. in conducting our review we looked at the many concerns raised by the gao and members of congress including those on this committee about the census bureau's cost estimate. i'm aware of the funding statistics presented by ranking

member cummings and can assure you the $760 billion budget shortfall and its impact is encompassed with our new lifecycle projections. we will hear testimony from both the gao and commerce ig about their concerns prior to thousand 15 lifecycle cost estimate. for example, the gao estimated the the it budget would be at least $4.8 billion. our estimate is $4.96 billion, so $160 million over the floor the gao estimated would be the correct number rather than the previously used one. i shared her concern and their testimony will underline the need for justification for the

new lifecycle cost estimate. ga all -- gao points out there are 43 technology systems involved in the 2018 end to end test. only four of those systems had completed development and integration tests as of august 2017. of the 39, a portion of the functionality has been deployed in the 2018 end to end test but that is not yet true for the remaining 18. with so much still under development, the need for a contingency to address possible overruns seems self-evident. these are just a few examples. the ig provided another example

regarding address candidacy where they found it would cost at least three times the amount for census bureau estimated in 2015. we agree with that and we have incorporated that into our numbers. on the whole we found the prior administration provided congress and the public with overly optimistic assessments of both the ease of implementing new technologies and the cost savings they would provide. these issues were undoubtedly compounded by initial appropriation constraints. procurement decisions created a series of silos most of which had a small firm general contractor and a number of

major firms as subcontractors. a very complex structure. many segments were on time and material spaces and in my view the most dangerous form of a contract. the administration failed to follow basic management practices like using certified cost estimators and checking guesstimates against actual costs. we have addressed these values. with this new lifecycle cost estimate. we identified key areas where census program it would likely create costs above the 2015 estimate. those include declining self response rates, increased public concerns about privacy and cybersecurity, high levels of mistrust of the federal

government, tightening in the labor market since 2010, development and integration of new technologies and the complex array of contractors and subcontractors. this leads us to project revised lifecycle cost estimate of $15.6 billion. the increase includes contingency funding of $1.2 billion to address additional potential risks and associated challenges such as rational disasters, potentially even lower self response rates. the difficulty and cost of hiring 500,000 temporary workers in a tight labor market

and the complexity of developing and integrating multiple information technology innovations. over 80% of projected spending, over 80% will occur post 2018. it is a very back ended expenditure as congressman cummings pointed out in his chart. this ramp up in the future brings its own challenges to implementation and cost control. our reserve, a contingency of 10% of the post-2018 on top of cost estimate level will be managed at the secretarial level and use only if unforeseen developments occur.

10% contingency is commonplace, estimating complex private-sector projects. government tends not to set up reserves but instead funds overruns in subsequent appropriation requests. our new $15.6 billion, 2020 census lifecycle cost estimate includes a request for a $187 million adjustment for fiscal year 2018. we are working with the house and senate appropriators on this request. these funds would allow us to make a significant course correction to keep crucial programs on track and provide much-needed financial oversight and better management of the

census bureau. the census bureau and department of commerce will be held accountable to manage the 2020 census efficiently while maintaining the highest quality. the undersecretary for economic affairs, karen kelly, has been in place since late august and has direct oversight over the 2020 census, the census bureau and the bureau of economic analysis. she brings over three decades of management experience in the financial investment sector, with experience managing people, strategy, operations, public relations and leadership across the globe, seasoned experience senior management is also in place at the census

bureau itself. ron jarman performing nonexclusive duties and functions of the director of the census bureau and performing the nonexclusive duties and functions of the deputy director and chief operating officer of the us census bureau jointly leading the census and each has nearly three decades of experience. congress conducts weekly, 2020 census oversight reviews and will require metric tracking and program execution status on a real-time basis. also monthly meetings with census program managers and 20/20 leadership team headed by undersecretary kelly to review

issues related to the program's budget, scope, schedule and risks. these management meetings include detailed reviews of the evolving budget and lifecycle cost estimates for the 2020 census. the result of these meetings are reported directly to me by undersecretary kelly and her staff and we maintain free flow of access and information so that i can personally oversee the progress of the decennial. we are just 30 months from the 2020 census. so many challenges ahead. they are urgently needed. with the changes i have detailed, with additional resources i am confident we will have a full, fair and

accurate census but look forward to working with this committee, your staff and the rest of congress over months and years to come and thank the committee for the opportunity to come before you today, look forward to the question. >> the gentleman from tennessee, mister duncan. >> thank you for being with us. many of us at the time of the last census received complaints in the community survey, some privacy advocates were very critical and our colleague, tadpole, let it charge against this calling that survey, quote, and unnecessary unwarranted government intrusion, led a charge to make

the survey voluntary. and we can't get as many complaints this time as we did the last time. and many states are going to be trying to do everything they can to make sure they don't lose a member of congress or they pick up one, many federal appropriations based on population. are steps being taken, no states can inflate their populations in those with the goal of getting into a congressional seat. >> those are two questions i will try to answer in sequence. most valuable data products,

widely used in the business community to make decisions. voluntary survey would have negative affect on the reliability of the data. especially rural and small communities. at the same time, for ways to reduce the burden on respondent such as removing questions by using other data sources including information people have provided government in a different form. we are trying to reduce the scope of the burden as a means of making it easier for people. in terms of trying to assure we help everyone we have made a number of changes. first of all the communications

budget for the 2010 decennial census was $350 billion and that included outreach to run primary english teak -- english-speaking communities. we increased that budget to $500 million this year, that equates to an inflation adjusted $420 million that would have been in 2010. we had a material step up in the communications budget. in the 2010-2020 census, about half of that will go for our partnerships, various faith-based and other community organizations that help us get to the more difficult to enumerate parts of the population. we are keenly aware of the need

for even greater effort than was used before. finally we have added internet response as an additional tool to make it more convenient for that portion of the population that likes to use internet. we made a number of very specific ratifications all of which we believe will lead to the end result of more accurate and more complete set. >> in maryland. >> to spend less than the 2000 census, not ingesting for

inflation, since that time congress starving the bureau of funds less then the census bureau requested in every single fiscal year since 2012. this must have led to investments in different technologies. is that right? does it affect your investment in new technologies? >> the expenditures if i understand correctly are the ones prior to the present period. i have seen the chart and it roughly parallels even though the distribution year-by-year is in the same, the 2021 roughly parallels the expenditures for the 2010 census. the 2020 census is back ended

as were the prior ones but even more so because of the nature of the technological changes we are making. in the supplemental material filed after today, fill in the blank for you and show how our projected budget for 2020 would compare both with the fluctuations, amplitudes and you will see we are coming to a significant higher absolute level and they have been. >> previous decades, and larger funding increases ramps up towards census day, the trump

administration did not recognize that. 8% increase in funding for the bureau in fiscal year 2018. >> i believe so and that was based on the information he then had from prior management of census. >> we presented the committee with data last week showing this level of funding is inadequate and could impair our ability to provide fair, and accurate census. you are now requesting additional $187 million to make up for what the department termed a quote back shortfall in the budget request for fiscal year 2018. is that correct? >> yes, sir.

i believe we do need $187 million more than was in. >> supports that request, omb supports that request. you also informed the committee last week, the commerce department and census bureau believe you need more than that amount for all eventualities, fate hundred $50 million to cover contingencies that almost certainly will occur. but omb denied that, right? for other contingencies. >> i don't know they took position on the lifecycle cost. the urgency in the budget had been getting their support for

immediate request for additional appropriation. >> you know that came to the attention of mister mulvaney. >> it was brought to his attention by me. >> running out of time. the president is aware of an additional $50 million. >> i have not personally spoken to the president about it. >> what will happen if problems arise? how you obtain funding to cover them? >> we believe $187 billion will cover us from fiscal year 2018, we are reasonably comfortable with that. the bigger risk comes after 2018 because that is when 80% of the money will be spent.

2018 is pretty high visibility, pretty well within the hand post that period for contingency. >> chairman yields back. >> thank you, mister chairman. the bureau's contracts, i asked the question about the contracts, time and material contracts means they are limited for labor hours and costs. what was the strong thinking behind this contract? >> i have no idea. i have been less worried about exploring the past then trying to make sure we get around the future. >> what can the department do to manage these contracts more effectively?

>> we are interacting aggressively with the contractors. that is about all we can do. many of the contracts have very severe cancellation provisions and would be very disruptive to cancel at this stage of the game in any event. we have to manage through the process pretty much with the existing group of contracting parties but the intensity of interaction i can promise you is more than it had been. >> one more quick question, with respect -- are you concerned certain product and systems may not be delivered in time for testing during the 2018 end to end test? >> so far as we can tell the ones that are incomplete are reasonably unscheduled but if

those would last 10%, that is the devil like in any contract, the devil is in the details. >> thank you. it is an honor to have you here. i'm a big fan of yours when you were on cnbc, i yelled back, mister chairman. >> the chairman yields back. >> secretary from the great state and city of new york, i would like to welcome you here today, thank you for your public service. in your comments, you seem very committed to getting an accurate count which is mandated in the constitution but if you look at the spending levels, you are new in this job but for the census for 2020, the red line is way behind, the prior census from 2010, 201990,

as a businessman. >> could you raise the chart a bit? i can't see the whole thing. >> let's take it down and give you one of these. this is a chart we have on the ratio of census budget to year one of the decade but in any event it is hard to achieve the results without spending so i certainly support your spending request of $187 million. it is a good start but not good enough. you have a lot to catch up on, several dress rehearsals were canceled, foreign language rehearsal canceled, this is a major undertaking, one of the major undertakings of our country and if we don't have a good system we don't have good data for government policy or business policy. i want to know what you are doing to address the immediate

problem. as i understand it your increase of $187 million is basically for it work but in the 1990 census was called a failure because of a declining response rate. in the 2000 census, the congress allocated more money and called for a paid advertising campaign and for partnership outreach in all types of efforts to build up

anticipation, in 202010 we were successfully increasing the public response rate it was critical in keeping the expensive follow-up method you have to do so that is very important. i read recently in the washington post that the ad contract congress mandated has not been funded yet and partnerships of not been fully filled and the census has to start early to be successful, so these partnerships with communities like the naacp, hispanics, lgbt communities, you have to fill this contract and they haven't been filled with people yet as i

understand. nowhere in your testimony or documents do you propose fully funding the contract as originally planned let alone increasing the contract to start addressing and tracking these cost drivers you identified in public outreach. i'm supporting your efforts. i introduced the 20/20 american census investment act which would provide more funding for the census and mandate that funds be used on promotion, education outreach, in addition to the it that you are addressing. i would like to hear your comments on whether or not you will be funding -- i want to thank chairman cummings or ranking member cummings for supporting the bill and efforts to increase funding for it.

to you feel you will be funding the ad contract, the partnership outreach? to get participation in the census. >> thank you, i am not used to being granted more money than i asked for. that is a new experience for me. >> we want to get an accurate account. as you look at this chart we are behind spending in prior censuses in this decade. >> as i said to ranking member cummings, we believe the cumulative shortfall of $176 million and its impact on the overall program is made up for by the funding we requested in the total lifecycle. >> will you be investing in

promotion and education? >> we are investing in all sorts of things. our communications budget will be $500 million as opposed to the 350 spent in 2010. inflation adjusting the 2010 would bring it to 420. we are running $80 million more than an inflation adjusted basis than the 2020. in addition -- >> my time is almost up. the economic survey, is that ontrack? >> i so so far. that is not the most urgent part of the problem of communications. what we are doing is several things. we have more ways people can respond than ever before. we are doing more language training than before.

we have introduced telephone call centers. they indicate the ability of operating in ten languages naturally including spanish. our population is more diverse than that. the call centers will be able to do up to ten languages. in addition, the census enumerators will be able to provide support in multiple languages. we are trying to match the characteristics of the enumerators to the characteristics of the population in the area they will be surveying. as to the partnerships we certainly agree that those are a big key. those will include national organizations, state and local governments, churches and other

faith-based organizations, legal aid centers and other support mechanisms. we are trying to work with all of those and the partnership program, approximately half of the 500 million budget for communication, about 248 million who go for the partnerships. we believe we are dealing adequately with the need for those. >> thank you very much. >> and the lady of that, gentleman from texas. >> i'm over here in the corner, other direction. got a couple questions for you. it seems to me that the census bureau has this attitude that we have got to do it ourselves, and aren't looking to products that are already out there or

companies you might be able to contract with to save some money. it is my understanding the census bureau received numerous recommendations from the gao and scientific advisory committee as well as the private sector, how the census bureau could use commercial mapping, geographic information technology which would yield a large cost saving and increase productivity but to date census bureau doesn't appear to recognize or implement any of these recommendations. we haven't been looking at things like that. >> there were many private organizations contacted in the original process. i can't talk about the process by which they made the final

selections they did, basically occurred years before but we are dealing with the contract that we do have. we believe these are qualified people and that we will be able to get the job done if we get the additional funding we have requested. in terms of lessons learned. one of the things we intend to do as we go along is careful track of further improvements that should be made in subsequent censuses. with only 30 months between now and the 2020 decennial, making radical changes would guarantee we didn't get it right. >> i'm concerned about the path we are on but if you could get me a list of some of these proposals and recommendations i would like to take a look.

we do the census every ten years and very few computer systems have an effect of life be on ten years. we will gear up with a new computer system in 2020 and here we are in 2030 sitting here doing the exact same thing when it seems to me you could get with a cloud service that is secure. we have got those that the government uses now and get an application written for people's phone, this seems a lot simpler and cloud service we would use to buy the computers, size from the way they need to be sized and when we are done with feeding massive capacity, they are there for other clients. that makes economic sense rather than the government added to the we have got to do everything ourselves especially in it where i think the

government really struggles, damaged by some of our own contracting laws. it seems this is a great opportunity to look for some cost savings. something i brought up to your predecessor several years ago that fell on deaf ears. >> we do believe the cost will come in the low inflation adjusted cost of the 2010 census notwithstanding the overruns. my predecessors estimated the inflation-adjusted cost for 2020 using the 2010 methodology would have been $17.5 billion. we are more than $1 billion below that despite problems that have occurred. second coming in terms of using existing databases we do what we can. for example making a lot of use

of administrative workers, postal system records, local irs, social security records to check ids, the truth is americans are very mobile population something like 12%, relocate each year. >> i want to make a comment, the postal service is an unused resource, more than they are currently using, postal service visit almost every american residence six times a week, there is a great resource that may be underutilized. >> using the postal service, the irs, social security,

medicaid, medicare records, indian health service, every database we can imagine, because we really are trying to make sure we count everyone one time. >> the gentleman from missouri is recognized. >> thank you, secretary ross, from the top, for your -- >> i don't know the geography of the committee well enough. >> i understand. the census is about three things, money information and power. no community or states wins if we fail to get this right and our nation will be the ultimate loser. as you all know our country is

more mobile and diverse than ever before and in some parts of our nation residents have a high level of distrust for the federal government which makes them less inclined to take part in the census. that is true in urban communities in st. louis which always struggles with lower self reporting response. generally across the nation african americans, latinos, new immigrants and the poor are at a much greater risk. a costly mistake paid for for an entire decade. underfunding the census 2020 outreach programs, hard to count groups is such an

enormous bipartisan threat to every state and every district that we represent. i guess i want to start the line of questioning following up on ms. maloney's questioning, hard to count, hard to reach audiences communications contract. must receive the full attention of your administration. i say that because the components of the communications contract needs to be made full partners in the communications effort because that is where the challenge

will be on how we get to those hard to count populations. if i could just hear your concerns about that and where we are going. >> the entity overall in charge of communications was involved in the 2020 decennial. first of all we have the institutional knowledge they gained by their role ten years ago. second, they have retained something like 15 individual firms that specialize in different forms of media or different ethnic groups. further, i may not have made it clear in what i said before, but $500 million of

communications is in addition to the $248 million for the partnership so altogether we are spending $748 million on this outreach initiative. when you consider how many total households there are, quite a bit through household and if you take out of that, the ones that are easy responders, spending an enormous amount that are difficult to reach. that is quite a deliberate thing. we have been conservative in our estimate to give voluntary response. last census with the 63.5%. we assume that may drop to 55%.

we assume despite massive communications we assume a higher percentage of the population will still have foot soldiers clogging the ringing doorknob. >> the partnership program calls for testing of communications methods. all communications, that is being counseled for 2018 so how will these elements be tested? >> it has been postponed. >> some time in the calendar year it will be tested. i thank you for that response and yield back. >> the gentleman from georgia. >> appreciate you being here, mister secretary. who is responsible for delivering these products on time? >> the census euro is responsible and they have a

series of contracting parties. they are probably 100 individual entities. somebody has to be responsible the product is delivered. >> ultimately me and karen kelly, as the oversight of census. >> why are the product not delivered on time? what is the hold up? for the delivery of the modernization products, we are going end to end and don't have what we need. >> the original estimate to how long it would take and how much it would cost to get to those products were flawed so some of it may be contractor not performing as well. i can't judge what went on prior to the present.

but i think it is a combination of over optimism on the part of census and some errors on the part of the contractor. >> is there any accountability, any penalty for not delivering on time what was promised to be delivered? >> each contract is different but when you are in a time and materials contract the concept of penalty doesn't work. >> we have ambitious, we hope to deliver, we won't deliver, there is no accountability. >> not as much as there would have been, if there were fixed contract but that is not the situation we are facing. >> when can we possibly guess under these circumstances that we will receive what we have?

>> we have given more than what we think is a guess. we have done more work in the field, lots of analysis of what has been done thus far and analysis of the contract. we believe this budget, the way we have proposed, will deliver the product we need. >> i don't share your optimism. i don't see how you can either with the situation we are dealing with. you look at cybersecurity, how confident are you that this information gathered with it that hasn't been tested will be secure? >> and today's world you can't be sure of anything in terms of cybersecurity but i can assure you the systems we are putting in our more robust than anything that has ever been done before in census. >> last year the cio made a

similar statement, they had confidence the cybersecurity was intact and the system having been tested, how can you have that confidence on something we know nothing about? >> i did not say there would be confidence it was impenetrable but a lot more robust than was true. >> how do you know that is my question? >> discussions with people doing the implementation of it. having said that i am not aware there is any system that is flawless. >> you are giving great confidence to us about cybersecurity based on conversation, conversations not enough to give us assurance this information is protected. >> what i'm telling you is the systems themselves will be far more robust than anything census has used before. >> we have not seen those

systems. >> they are works in progress. >> it is based on conversation, not tests. this is all part of the problem. we have promises that are never delivered on and now cybersecurity promises that most robust thing we have seen, based on conversation, not on tests. looks like we are going back to the same thing that happened in 2010 when we were promised modernization but it was a faulty program so it is a paper-based census. i don't see how we can proceed any further with what we are dealing with. >> it is impossible to test something that doesn't yet exist. >> my point precisely. >> 80% of the spending will be toast 2018 but let me describe

to you what it is we are doing. all the data on the census systems are being protected, applying rigorous security protocols to the perimeter of the census network. the census has an existing network. we are running rigorous protocols to that. we are putting in routers and firewalls and also have our own and prevention systems in place, census euro now has an unprecedented level of engagement with the best resources elsewhere in the government working closely with cyber security experts across the department including an isc which is part of commerce, the

federal cios, department of homeland security and intelligence and law enforcement communities, putting a new level of cyber security federal management in place for the 2020 census over and above what was done in 2010. >> our time is gone. we went to respect the time. talking points to me, not much beyond that. >> you can test something that isn't in place. >> the gentleman yields back. the gentle lady from the virgin islands. ..

that are potentially in place to be able to do that. we know that millions of americans have been displaced by hurricanes harvey, irma and maria in the states of texas, florida, and the arizona puerto rico and the virgin islands. as the communities work tirelessly to rebuild large-scale changes are largely to occur that i would assume would greatly impact the 2020 census. if we've learned anything from hurricane katrina in 2005, it is that recovery from large scale natural disasters take time and often quite a lot of that time. 2011 report by the leadership conference on civil and human rights illustrated challenges to enumeration activities still exist in the gulf region in 2010, almost five years after katrina took place.

since the report highlighted issues such as large numbers of individuals and temporary housing, informal housing arrangements, high rates of vacant units come significant changes in both new housing stock and population in those affected areas. mr. secretary, do you agree the bureau is likely to encounter many if not all of these challenges due to the destruction caused by this year's hurricanes, and perhaps future hurricanes before the census takes place? >> well, you are surely right in pointing out issues. here's what we are doing. during the 2010 census, they changed the operations in the areas that were hit by katrina and rita. basically what it involved was putting more people on the ground to deliver questionnaires directly to housing units, then updating the addresses while

doing so. we can do something similar in 2020, as needed. if it does turn out that we need that, it will cost more money, but we have tried to allow an amount for these types of contingencies, provided that the natural disasters are not more extreme than we have experienced in the past. in addition, we have a particular problem in the u.s. virgin islands, and for that matter in puerto rico. we conducted a census in those territories while working directly with territorial governments to build and implement the operations that best fit their respective situations. we are currently developing our approach for the 2020 with the government of virgin islands and will ensure that the operations take into account the impact of

the recent hurricanes. >> excuse me. when you say particular challenges, what have you seen in the past as being uniqueness of trying to find census data in areas like puerto rico and the virgin islands, besides the obvious which is we don't have street addresses? >> right. as you know the census has confronted those basic problems over and over and over. so every time that they had gone through it they learned something from the process. the difficulty we have now is that issue with residences have been terribly compounded by the awful destruction that occurred. so the magnitude of the task has become a greater that it ever had been. >> and will your budget reflect the magnitude of these

hurricanes? >> we have tried our best to do so. >> avenue in the past you've done a a good job of not just partnering with the local governments, but with community organizations as well, that they've been somewhat boots on the ground -- >> absolutely, absolutely. and that's why we've allocated $248 million for partnerships with local community organizations. >> okay. thank you. thank you very much. i yield back. >> the gentleman from alabama is recognized. >> thank you, mr. chairman. thank you, mr. secretary, for being here today. i understand some of the frustration that my colleagues have expressed, and give an example, the gao and the office of inspector general have cited a a number of issues where the census bureau, that isolate you

inherited had to be totally truthful about it, but the frustration is that these don't appear to have been adequately addressed. for instance, the gao included the 2020 census on its high risk list for 2017. in fact, over the last three years they have issued 30 recommendations, but as of january of this year only six have been implemented. are you familiar with this list? >> well, we have been working closely with the gao. you will be hearing from them a bit later. and we are cooperating with them in their audit of -- >> my question, mr. secretary, we only have five minutes and you want to give you an opportunity to answer this as a yes or no would be sufficient. are you familiar with this list of 30 recommendations? >> i couldn't hear you. >> are you simply with his list from the gao, the 30

recommendations? >> sure. >> and you are aware that only six have been implemented come have been followed up on as of january of this year? >> as of january that was true. i believe there's been some for the progress since then and there will be continuing progress. >> i think it's extremely important that, that when the gao or the oig make recommendations when you have findings of issues, that federal agencies, federal departments, act on them. i mean, it's frustrating to the folks were doing the work. it's frustrating to us in our oversight capacity, and it's not just the census bureau. this has occurred and other agencies but there's frustration that we identify problems and they don't get addressed. my question to you is, is this a top priority, mid-level? how serious are you taking these? >> well, it certainly is a top

priority, and i think you'll find the oig testimony pretty well mirrors the top management challenges report that we put together. what happened prior to january of this year i can't account for. i was confirmed that you were 28. >> i realize that and i made that clear that you inherited most of these problems. but the thing that we want to know is that under your leadership, that this this is n seriously. we've known this -- the gao begins questioning that in 2008. >> it certainly is, and my testimony today focus on a lot of it. one of their recommendations is the bureau needs to manage the risk of implementing innovation. you heard my testimony. i totally agree with that, and we are much more heavy-handed in

the management then that had been before. second, the bureau continue to face challenges in implementing and securing key i.t. systems. 100% true, for sure. and the 2018 into in task will probably reveal some more deficiencies that will have to deal with. >> mcauley from george expressed frustration with the process and i appreciate your response that you can't test a system until the system is a place. the bureau has had a number of problems in terms of failing to adequately test systems and products. and when they do test them they are not using the results to inform their decision-making. >> well, we will. >> that's comforting to know, and i am certain under your leadership that that will happen. but that's one of the frustrations again that we have is that you have the gao, the

oig, this committee making recommendations. it's not just the commerce to department. we want to make sure this is such a serious issue, we want to make sure that systems are tested and the results are used in decision-making. mr. chairman, i appreciate the opportunity to submit questions and i yield back. >> and the i.t. part was a very key one in the gao recommendations. and as i had mentioned earlier, our budget is more than what they felt was the minimum required. they required thinking, as opposed to i think it was 36, that the former administration had estimated, think they recommended 4.8 or more. more. we came into 4.960. so we are not ignoring any of those recommendations not by a

longshot. >> the gentleman from alabama yields back. the gentlelady from florida is recognized. >> thank you so much, mr. chair, thank you so much for being with us today. you've already talked about undercounted populations. but this issue has been in the news recently because of a lawsuit filed by the naacp, which it filed in part because of concerns about the undercount of communities of color, junk children, home renters, room residence. i am concerned that the undercount could be exacerbated in 2020 if minority groups are less inclined to share personal information with the government due to the climate of mistrust created by this administrations immigration policy. secretary ross, can you assure

immigrant families and others bad word, but marginalize groups, that the data will be kept confidential and not used to pursue any law enforcement or immigration action? >> i believe it would be illegal for those data to be used for other purposes. so unless someone commits a crime, i don't think that's going to happen. but on the naacp lawsuit, they filed a freedom of information act request with us on the 29th of june. >> secretary ross, , excuse me just a second. you are saying if the information was used for law enforcement purposes or for immigration policies, that that would be illegal? >> my understanding is that the information about specific

individuals is not to be used for any other purpose. >> and using it would be illegal -- i'm just trying to understand that's what you're saying. >> i'm not a lawyer so i can't parse that, but as i understand it, it's not to be used for any other purpose. >> okay. one way you've also spoke about partnership programs, and certainly i believe that additional programs of that nature may lessen some of the concern. how many partnership specialist have you hired and how many do you anticipate hiring by census date? >> quite a few. let me see if i can get you the exact number. >> okay. and while you are looking for that, i think my colleague, mr. clay, talk about testing a new process, particularly -- >> forty, 40 people. >> and how many do you anticipate hiring by census day, or is that the total number?

>> forty i believe is the total number. >> okay. >> to date. >> thank you very much. any anticipated numbers by census date at this point you are not sure? >> we will see how much comes in. a lot of community organizations are well aware of the census, have participated in it before and are signing up themselves. so depending on whether we get a good flow, , consistent with improving it over the last time, that will determine how many people are needed. >> okay, thank you for that. i know you also spoke about testing of new procedures. i'd like to talk particularly about in rural allen and tribal areas, and also tying those, those procedures to the budget, constraints that we currently face. secretary ross, what are you doing to ensure that these new

processes will work smoothly in rural ireland and tribal areas? >> well, tribal areas are one specific segments that are being tested so that we can get real round results come sing how the procedures that we've introduced work. >> okay. and for rural and island? >> saying. >> thank you very much for that. on the census website, it says and a quote we promise that we would use every technology, statistical and secreta procedure at our disposal to protect your information, unquote. could you talk, secretary ross, a little bit about have adequate or inadequate funding might affect that promise? >> well, we believe that the funding we have requested will provide us the resources we need

for that purpose. >> and if you do not get the funding that you requested, are you still able to fulfill the promise that is on your website? >> we will do the best we can with the resources we get, but we think that the optimal amount for the proper job is the amount we have requested. >> thank you, secretary ross. mr. chairman, i yield back. thank you. >> the gentleman from iowa is recognized. >> thank you, chairman gowdy. secretary ross, thank you for being here today. 15.6 billion i guess is the latest estimate in iowa. that's a lot of money. >> it is to me, , too, sir. >> 125 million households i think there are, so the math, if i'm correct is about $125 per household. >> just about right. >> do what we want to do.

does that strike you as a large number per household? >> it is a larger number, and it's even a larger number when you consider that the vast, bulk of that is spent on getting the last few million in. because the one to respond to the initial mail the document or respond over internet, the cost of those is very small. it's the more difficult to enumerate people that are really the most expensive to get you. the last few million people are going to cost infinitely more than the first few million. thus far that is why the response rate is a critical. >> your private sector guy, i'm a private-sector business person as well. can you in 60 seconds let me know how much, are we utilizing the private sector? i'm thinking of, for example, google. probably didn't spend 15 billion

to map the entire united states visually. are we utilizing the technology and resources available in the private sector to the extent possible? >> i believe so. i believe so. >> can you expand on that just briefly? especially the technology piece of this. >> right. well, where the google type technology is helpful is in locating physical structures. it doesn't really go within the structure to the individual person. so our task is first to identify the physical structures that have residence in them, whatever the nature, whatever the formality or informality of the living arrangement. but then the real test is to get inside them and find out exactly how many people are there. so that's where the self responses are very helpful, and

we are providing for internet self response. we're providing call centers to make it easy. everybody's going to be getting mailings. and then finally as a last resort we would use the primitive thing of a guy knocking on the door. >> and speaking of, i'm glad you said i guy knocking on the door. back in my district in iowa i have delivered the mail with postal carriers. one of them brought and i do to me. they said why don't we do, why don't they use us more? he said, for example, i know exactly on my route how many people live in each house. i know their names. he says i know a lot about them. why don't we utilize us? that's a great question. i would like to post that to you. >> well, we do. our fundamental database is the postal system. no question. but because people move, people die, people, all kinds of things happen. something like 12% of the

population relocates during the course of a calendar year. and within the more difficult, to enumerate portion, there's probably even more mobility and there is say in single-family home in a little town. it's very, very complicated but we do begin with the postal system records, for sure. >> so use their database. can we also utilize the man knocking on the door? can we utilize the postal carrier for that? >> a man knocking on the door is the last resort. only when all the more efficient mechanisms have already been used. >> looking at the 15.6 billion and i thought, mr. secretary, if you and i formed a company i guess we could do it for 8 billion, save the government seven and you and i would put a few dollars at our pockets and guessing. is there any truth to that? >> well, i don't know.

we are where we are, and we do think under the circumstances we really need the 15.6. >> thank you very much. thanks for being here, mr. secretary. i yield back. >> the gentleman from pennsylvania is recognized. >> thank you, mr. chairman. chairman gowdy, want to express my appreciation for this hearing. people don't think about the census that much, but it is such an important thing. not just because it's constitutionally mandated that we need to get it right. we need to get it right so our american federal government can work better for everybody. we need an accurate since his. we need to know whether people are. we we need to know where the jobs are. we need to know where the poverty is. we need to know how the economy is changing. we need to know what the government needs to do, and

maybe needs to stop doing, and where these things need to happen. so so much of policy depends on an accurate census. we need to get it right. secretary ross, it's good to see it again. i saw you in maine when you came and testified before the house appropriations committee -- in may -- and want to touch on some areas that we do have concerns about. i think all of us in this room share the concern, we need to get this census right. we want to learn from past mistakes. we don't want to be, we don't want to have rose-colored glasses, and really unjustified hopes. we need to be very practical and pragmatic about this. and i think you with that kind of person, secretary ross, so want to touch on a few things. first, warning signs of trouble. i mean, may 3 was when census director thompson appeared

before my subcommittee on appropriations. we asked him hard questions, much like the ones you are getting your today, secretary ross, and he resigned the next week. that's a a warning sign of troe to me. and he hasn't been replaced. first question, are their immediate plans to replace the census director? >> yes. we had actually come up with a nominee, and then that fell way through the vetting process so we are actively trying to recruit. but i have very good confidence in the team that we have assembled right now, namely the two people who collectively have some 30 years each of experience are there. they are functioning every day. and karen dunn kelly is now providing very direct oversight of them, plus i am being very personally involved and we have

the whole series of oversight meetings, meetings with the contractors. we are all over this thing as best we can be. so it's not to say that we don't want to have a permanent approved census director, we do and we will. it's not an easy task to fill because you need someone who has subject matter command and management experience to deal with the enormously challenging task of hiring a half 1 million temporary workers, getting them out there, getting them productive and getting this big technological change introduced. >> you do have a lot of other things that demand your attention, secretary , secretari urge you to devote a maximum effort into finding that person to be a strong, active, knowledgeable census director. let's get that done quickly. >> we are trying our very best.

>> we need to get this census right, , and one of the red flas is we know there are underreported communities, and a number of them have been mentioned by my colleagues on both sides of the island today. but one of them i've noticed is the lgbt community. the art historically underreported. we know that they face higher levels of poverty, particularly older lgbt people, that lgbt young people are more likely to be homeless, making it all the more likely that they will be not counted. and, in fact, the census bureau itself has recognized that the lgbt community is a hard to count population, yet the bureau has declined to put questions about sexual orientation and gender identity into the census. secretary, did you determine

there was no federal need for this information? >> no. there were hearings held about it. there were requests put in, and it was concluded that that particular set of questions did not meet the requirements for being put in. one of the problems with adding questions is it reduces response rates. it may seem counterintuitive, but the more things you ask in those forms, the less likely you are to get them in. it's a balancing act between more information and fewer responses. >> secretary ross, my time is up but i want to associate myself with the hard questioning you got from my colleague, congressman heist of the georgia. i urge you to devote a great deal of attention to the online, the self reporting, increasing

public confidence in the cybersecurity of the system. because of you said yourself today, that's the cheapest way we get this job done with the self reporting, with the online work. and i urge you to devote a lot of attention to increasing the public's trust in the online system and giving them good reason to have trust in that system. and with that i yield back. >> we share your concern. >> thank you. >> the gentleman from california is recognized. >> sorry about that, the gentleman from texas. i apologize. >> i do want to say this in his defense. he was up several times ago can you let others go. i apologize. you are up and then the gentleman from california. >> thank you, chairman. it's great to know that there is sweet attention on this important issue. you have a very important portfolio but it's great that

you are involved in providing leadership for this. i know you have inherited i think the technical term is a mess. and i recognize, and i appreciate your comments about how to prevent the next census from being in a similar position, so i appreciate that. when some of your predecessors were here last time, there was a debate about whether the census bureau should develop a widget that uses the telecommunications backbone that the enumerator can put data in. and the question was that widget already exist. it's called a smart phone. and i'm glad to know that the census bureau decided to use i believe smart phones. my question around that is, as we discussed or have plans come have you and the cio of the census bureau had some conversations around the

security of whatever device is going to be used by the enumerators? >> yes. we've had consultation not only within the department, but with homeland security, with fbi, with the intelligence community, with the overall cios and the government, and in commerce. so we're building in the best firewalls and routers that we possibly can. but having said that there is probably some 13-year-old kid in a garage garage somewhere trying to figure out how to penetrate it. so in today's world you do your best but you can't be 100% sure, unfortunately. >> had to begin with the assumption of breach. i did this for a living. i never not got in. but the fact that there is again a focus from your office on the security of these systems, and i know that there is 43 systems

that have to be delivered for end-to-end testing in 2018. and only for those have been delivered. i know this conversation on that. -- i know there's conversations on that. my concern is the bureau is going to have to get the authorizations to operate, the hbo's for each one of these systems. i know some of these, , sectionf these 43 systems already have ato's but they will need to get new ones because the system is evolving. and i'm curious in the strategy that you and your i.t. team have discussed on ensuring that we have those ato's in time to protect those personally identifiable information, especially on those 33 systems that do have pii.

>> well, there's actually more than 50 systems that will be involved. there uses are to update and maintain the address list to enable people to respond via the internet, to identify the households that have not responded in order to send interviewers to them, to manage the work of the interviewers, then to recruit and pay the field workforce. those are just a few examples of the kinds of systems. systems also capture, process, and tabulate all of the data collected on the people living in the united states. we are monitoring the progress of all of the systems, and each one is at a different stage of completion, as we sit here. nd