1 00:00:02,480 --> 00:00:14,000 Speaker 1: Bloomberg Audio Studios, Podcasts, Radio News. 2 00:00:18,079 --> 00:00:21,599 Speaker 2: Hello and welcome to another episode of the Odd Lots Podcast. 3 00:00:21,680 --> 00:00:24,040 Speaker 3: I'm Joe Wisenthal and I'm Tracy Alloway. 4 00:00:24,360 --> 00:00:26,880 Speaker 2: Tracy, I've said this a bunch of times, and we've 5 00:00:26,920 --> 00:00:29,600 Speaker 2: sort of danced around this issue on the podcast before, 6 00:00:29,640 --> 00:00:31,080 Speaker 2: but I've said this a bunch of times. I'm sort 7 00:00:31,080 --> 00:00:33,640 Speaker 2: of fascinated by the degree to which we sort of 8 00:00:33,720 --> 00:00:36,800 Speaker 2: take numbers on the screen. For granted, Like you know, 9 00:00:36,920 --> 00:00:38,879 Speaker 2: we look at a price of a stock and it 10 00:00:38,960 --> 00:00:41,160 Speaker 2: exists on the screen, it's like, where did that come from? 11 00:00:41,159 --> 00:00:42,080 Speaker 4: How did that get there? 12 00:00:42,520 --> 00:00:45,000 Speaker 2: Or like the first Friday of the month, the job's 13 00:00:45,080 --> 00:00:47,680 Speaker 2: data flashes on the screen and we just start talking 14 00:00:47,920 --> 00:00:50,920 Speaker 2: about what the data said, what the unemployment rate was, 15 00:00:51,000 --> 00:00:53,360 Speaker 2: et cetera. But we don't really talk enough about like 16 00:00:53,400 --> 00:00:56,520 Speaker 2: what had to happen behind the scenes to get that 17 00:00:56,640 --> 00:00:57,880 Speaker 2: number to the screen, right. 18 00:00:58,000 --> 00:01:01,080 Speaker 3: I find this really interesting as well, because obviously there's 19 00:01:01,120 --> 00:01:04,040 Speaker 3: the data collection portion of this, Like, yeah, you have 20 00:01:04,120 --> 00:01:07,640 Speaker 3: to go out and talk to people for certain surveys, 21 00:01:07,680 --> 00:01:10,960 Speaker 3: certain data points, but also there are so many qualitative 22 00:01:11,120 --> 00:01:14,840 Speaker 3: and subjective adjustments that you can make to that data. So, 23 00:01:15,160 --> 00:01:19,560 Speaker 3: for instance, with CPI I didn't know that CPI waitings 24 00:01:19,600 --> 00:01:23,000 Speaker 3: are like different depending on what city you're in. So, 25 00:01:23,040 --> 00:01:26,039 Speaker 3: for instance, like food at home could matter a lot 26 00:01:26,120 --> 00:01:30,600 Speaker 3: more in I don't know, Minneapolis compared to Chicago. It's 27 00:01:30,680 --> 00:01:33,920 Speaker 3: really interesting. And there's also qualitative adjustments. Yeah, so if 28 00:01:33,959 --> 00:01:38,400 Speaker 3: your fridge gets Wi Fi connected, then that gets incorporated 29 00:01:38,480 --> 00:01:41,600 Speaker 3: into the price as well. So it's really interesting. 30 00:01:41,720 --> 00:01:44,880 Speaker 2: It's super interesting, and like I have a you know, 31 00:01:45,240 --> 00:01:47,920 Speaker 2: this is the issue. We don't talk about data collection, 32 00:01:48,480 --> 00:01:50,960 Speaker 2: but it's obviously some of the most interesting stuff there 33 00:01:51,080 --> 00:01:53,200 Speaker 2: is because of course, you know, things like food at 34 00:01:53,200 --> 00:01:55,360 Speaker 2: home and one city should be different than another city 35 00:01:55,400 --> 00:01:57,920 Speaker 2: if you really want to collect a basket, and so 36 00:01:58,120 --> 00:02:00,360 Speaker 2: means that you have is some of the most interesting 37 00:02:00,440 --> 00:02:05,080 Speaker 2: economics work being done anywhere in a realm that virtually 38 00:02:05,120 --> 00:02:07,880 Speaker 2: gets no attention. And when we talk about data collection, 39 00:02:08,280 --> 00:02:10,560 Speaker 2: you know, you hear it a lot in politics, right, 40 00:02:10,919 --> 00:02:13,680 Speaker 2: surveys have gotten you know, no one answers the phone. 41 00:02:14,480 --> 00:02:16,560 Speaker 2: It's you know, it's harder and harder to do high 42 00:02:16,680 --> 00:02:19,560 Speaker 2: quality surveys. I think you've written about this, like response 43 00:02:19,639 --> 00:02:22,280 Speaker 2: rates and just the actual cost of collection of all 44 00:02:22,280 --> 00:02:23,560 Speaker 2: this data is on the rise. 45 00:02:23,680 --> 00:02:25,880 Speaker 3: Yeah, so I have a bunch of thoughts on this. 46 00:02:26,080 --> 00:02:29,040 Speaker 3: I'll just say talking about the economic data might be 47 00:02:29,200 --> 00:02:32,960 Speaker 3: timely as well, because we know that Trump doesn't really 48 00:02:33,360 --> 00:02:36,560 Speaker 3: respect I guess a lot of official economic data. And 49 00:02:36,600 --> 00:02:39,520 Speaker 3: he's also trying to cut back on costs or funding 50 00:02:39,600 --> 00:02:42,880 Speaker 3: of a bunch of different government programs. So it's clear 51 00:02:43,040 --> 00:02:46,279 Speaker 3: that you know, entities like the Bureau of Labor Statistics 52 00:02:46,280 --> 00:02:49,320 Speaker 3: could potentially be in the crosshairs here. But you're right, 53 00:02:49,600 --> 00:02:53,480 Speaker 3: response rates lower, response rates have been happening for a 54 00:02:53,520 --> 00:02:56,800 Speaker 3: long time, and it's really easy to look up like 55 00:02:57,040 --> 00:02:59,400 Speaker 3: just how bad things have gotten. If you look at 56 00:02:59,400 --> 00:03:03,080 Speaker 3: the BLS website, for instance, look at response rates on 57 00:03:03,480 --> 00:03:07,399 Speaker 3: the housing portion of CPI that's gone from about seventy 58 00:03:07,440 --> 00:03:12,200 Speaker 3: percent back in twenty fifteen to just fifty seven percent. Now, wow, 59 00:03:12,480 --> 00:03:15,520 Speaker 3: response rates on jolts have gone from I think sixty 60 00:03:15,560 --> 00:03:20,399 Speaker 3: seven percent to just thirty percent, So that's pretty amazing. 61 00:03:20,480 --> 00:03:23,440 Speaker 3: And the other interesting thing here is it's not just 62 00:03:23,480 --> 00:03:27,280 Speaker 3: a US problem, right, Like there's no US exceptionalism here. 63 00:03:27,560 --> 00:03:30,240 Speaker 3: You see the same pattern in other economies like the 64 00:03:30,320 --> 00:03:33,880 Speaker 3: UK and New Zealand, and actually in the UK just 65 00:03:34,000 --> 00:03:37,320 Speaker 3: last month, the Office of National Statistics said it wasn't 66 00:03:37,360 --> 00:03:40,960 Speaker 3: going to publish PPI because of a data quality issue. 67 00:03:41,160 --> 00:03:44,000 Speaker 3: And also trade data had a problem because of errors 68 00:03:44,080 --> 00:03:47,520 Speaker 3: in the data provided by HM Revenue and Customs. And 69 00:03:47,560 --> 00:03:49,800 Speaker 3: now there's a task force to look at all of 70 00:03:49,840 --> 00:03:53,640 Speaker 3: these mistakes and data problems at the ONS. So something 71 00:03:53,640 --> 00:03:55,640 Speaker 3: that goes beyond the US here. 72 00:03:55,960 --> 00:03:58,000 Speaker 2: This is a really big deal, especially when you just 73 00:03:58,080 --> 00:04:02,400 Speaker 2: consider how much economic activity is based on being able 74 00:04:02,480 --> 00:04:05,520 Speaker 2: to look at high quality data and make decisions from it, 75 00:04:05,880 --> 00:04:07,920 Speaker 2: obviously in the market, but also in just sort of 76 00:04:07,920 --> 00:04:11,520 Speaker 2: the quote real economy, et cetera. Anyway, without further ado, 77 00:04:11,680 --> 00:04:15,360 Speaker 2: we really do have the perfect guest, someone who knows 78 00:04:15,400 --> 00:04:18,440 Speaker 2: a lot about how the sausage is made and why 79 00:04:18,480 --> 00:04:21,880 Speaker 2: making the sausage is getting more expensive. We are going 80 00:04:21,920 --> 00:04:24,440 Speaker 2: to be speaking to Bill Beach. He was most recently 81 00:04:24,480 --> 00:04:28,440 Speaker 2: the commissioner of the BLS, and he knows all about this. 82 00:04:28,520 --> 00:04:31,599 Speaker 2: He was the fifteenth commissioner at the BLS, does research 83 00:04:31,920 --> 00:04:35,680 Speaker 2: affiliated with the Economic Policy Innovation Center, and he's going 84 00:04:35,720 --> 00:04:37,640 Speaker 2: to talk to us about all of these issues. 85 00:04:37,680 --> 00:04:39,719 Speaker 4: Bill, thank you so much for coming on odd Lats. 86 00:04:40,520 --> 00:04:42,159 Speaker 5: Oh Man, it's just it's great to be with you 87 00:04:42,200 --> 00:04:44,559 Speaker 5: and Tracy. Thank you very much. It's a great topic. 88 00:04:44,800 --> 00:04:45,640 Speaker 4: It's a great topic. 89 00:04:46,160 --> 00:04:48,839 Speaker 2: I don't know like when the process begins, but it's like, 90 00:04:49,279 --> 00:04:52,200 Speaker 2: I get this number. It says what the jobs report, 91 00:04:52,360 --> 00:04:55,680 Speaker 2: it says how many jobs were created that month? How 92 00:04:55,680 --> 00:04:59,120 Speaker 2: did that number get onto my screen or ontobls dot gov? 93 00:05:00,040 --> 00:05:01,760 Speaker 5: And it gets there every month, Yeah. 94 00:05:01,720 --> 00:05:04,640 Speaker 2: Every month, every month. It's never missed except maybe once 95 00:05:04,680 --> 00:05:07,719 Speaker 2: send like a hurricane or so. Anyway, So to get. 96 00:05:07,560 --> 00:05:11,760 Speaker 5: That first, yeah, the first Friday report, the Jobs Report, 97 00:05:11,920 --> 00:05:14,599 Speaker 5: which comes out at eight thirty Eastern time on the 98 00:05:14,640 --> 00:05:18,000 Speaker 5: first Friday. It consists of two surveys. So let me 99 00:05:18,000 --> 00:05:19,640 Speaker 5: talk about the first one, which is the one you 100 00:05:19,760 --> 00:05:23,280 Speaker 5: just mentioned that the number of jobs. That is a 101 00:05:23,320 --> 00:05:28,040 Speaker 5: survey of about four hundred thousand firms out of eleven 102 00:05:28,080 --> 00:05:32,000 Speaker 5: point three million firms in the United States, about four 103 00:05:32,080 --> 00:05:36,240 Speaker 5: hundred thousand of them have agreed voluntarily to send in 104 00:05:36,760 --> 00:05:41,800 Speaker 5: a pretty complex survey response. Every month, they send that 105 00:05:41,839 --> 00:05:45,960 Speaker 5: survey to electronic collection centers. There's one in Chicago, there's 106 00:05:45,960 --> 00:05:51,000 Speaker 5: one in Atlanta. And that survey has to contain the 107 00:05:51,160 --> 00:05:54,680 Speaker 5: twelfth day of the month, if it is a work day, 108 00:05:55,279 --> 00:05:58,880 Speaker 5: or the closest workday to that twelfth of November, or whatever, 109 00:05:59,240 --> 00:06:01,039 Speaker 5: and the reason for that is we want to get 110 00:06:01,120 --> 00:06:06,400 Speaker 5: at least one pay period in the report. Okay, So 111 00:06:06,680 --> 00:06:11,800 Speaker 5: they submit that. That data continues to be collected through 112 00:06:11,920 --> 00:06:15,440 Speaker 5: almost the end of the month, not quite, but almost 113 00:06:15,480 --> 00:06:18,920 Speaker 5: the end of the month. It goes from regional offices, 114 00:06:19,080 --> 00:06:23,960 Speaker 5: these regional collection centers to the national headquarters in Washington, 115 00:06:24,440 --> 00:06:28,640 Speaker 5: where about forty people sometimes less, out of the two 116 00:06:28,680 --> 00:06:32,320 Speaker 5: thousand people that work there massage the data, which means 117 00:06:32,320 --> 00:06:35,640 Speaker 5: that they take the data which is in the survey. 118 00:06:35,720 --> 00:06:38,599 Speaker 5: It's a sample, it's just a fragment of the total population, 119 00:06:39,160 --> 00:06:43,520 Speaker 5: and they multiply the responses by what are called weights, 120 00:06:43,800 --> 00:06:47,640 Speaker 5: and that gives us the national number. So that number 121 00:06:47,760 --> 00:06:52,479 Speaker 5: is all done pretty much by Tuesday night, preceding the Friday, 122 00:06:54,120 --> 00:06:58,960 Speaker 5: and on Wednesday, the Commissioner gets to hear the data. 123 00:06:59,120 --> 00:07:03,160 Speaker 5: The Commissioner, by the way, plays no role whatsoever in 124 00:07:03,440 --> 00:07:08,680 Speaker 5: massaging the data or multiplying the survey results times the weights. 125 00:07:09,560 --> 00:07:13,200 Speaker 5: Then I would brief the White House. The White House 126 00:07:13,200 --> 00:07:17,160 Speaker 5: would then brief the Federal Reserve Ward chairman, the Secretary 127 00:07:17,200 --> 00:07:21,400 Speaker 5: of the Treasury on Thursday. That Thursday night, usually they're 128 00:07:21,440 --> 00:07:25,880 Speaker 5: sworn to secrecy, and then at eight o'clock in the morning. 129 00:07:25,920 --> 00:07:28,920 Speaker 5: I would walk over to the Department of Labor, because 130 00:07:29,000 --> 00:07:32,360 Speaker 5: blas's headquarters is about a twenty minute walk from the 131 00:07:32,360 --> 00:07:35,680 Speaker 5: Department of Labor. I would brief the Secretary of Labor 132 00:07:36,320 --> 00:07:39,920 Speaker 5: and for an hour from eight thirty to nine thirty, 133 00:07:40,240 --> 00:07:42,760 Speaker 5: the Secretary of Labor and his staff had to remain quiet, 134 00:07:43,440 --> 00:07:46,160 Speaker 5: and then at nine thirty they could answer questions of 135 00:07:46,200 --> 00:07:49,720 Speaker 5: the press. So that basically is the process for the 136 00:07:49,840 --> 00:07:54,040 Speaker 5: jobs report. There's another survey there. It interrupt me any time, 137 00:07:54,120 --> 00:07:57,240 Speaker 5: but that's where we get the unemployment raise, called the 138 00:07:57,280 --> 00:08:02,040 Speaker 5: Household Survey, and that's a survey of sixty thousand households 139 00:08:02,720 --> 00:08:06,280 Speaker 5: selected out of a sampling frame. We can talk about 140 00:08:06,280 --> 00:08:10,160 Speaker 5: that term to be representative of the entire United States 141 00:08:10,400 --> 00:08:13,000 Speaker 5: by demographic segments. You know how many people are in 142 00:08:13,040 --> 00:08:17,840 Speaker 5: the certain age group, male, female, the racial and ethnic compositions, 143 00:08:17,840 --> 00:08:22,520 Speaker 5: and geographic locations. So it's a complex sample. That sample 144 00:08:22,720 --> 00:08:26,720 Speaker 5: is done usually by Monday of the week prior to 145 00:08:27,880 --> 00:08:31,240 Speaker 5: the release on Friday, and then again I get to 146 00:08:31,240 --> 00:08:33,439 Speaker 5: see the results, so the commissioner gets to see the 147 00:08:33,480 --> 00:08:38,200 Speaker 5: results on Wednesday. So that's kind of the calendar. Both 148 00:08:38,240 --> 00:08:42,400 Speaker 5: the household Survey and the Employment survey cover the same 149 00:08:42,720 --> 00:08:45,959 Speaker 5: two week period in the month. Now there may be 150 00:08:46,000 --> 00:08:50,520 Speaker 5: a day or two switch there. Sometimes the household survey 151 00:08:50,600 --> 00:08:53,920 Speaker 5: has a few more days in it than the establishment survey, 152 00:08:53,960 --> 00:08:56,880 Speaker 5: but they all have that twelfth of the month because 153 00:08:56,880 --> 00:08:59,280 Speaker 5: we're trying to get the middle of the month. By 154 00:08:59,320 --> 00:09:01,720 Speaker 5: the way, that's really important because if you have a 155 00:09:01,800 --> 00:09:04,920 Speaker 5: natural disaster that happens prior to the twelfth or after 156 00:09:04,960 --> 00:09:07,920 Speaker 5: the surveys are closed, that does not affect the results, 157 00:09:08,000 --> 00:09:10,000 Speaker 5: even though you would think since it happened in the 158 00:09:10,040 --> 00:09:12,600 Speaker 5: month prior that you know, the hurricane might have affected 159 00:09:13,040 --> 00:09:14,240 Speaker 5: implument results. 160 00:09:29,840 --> 00:09:33,840 Speaker 3: So one thing you mentioned is companies, you know, voluntarily 161 00:09:34,080 --> 00:09:37,520 Speaker 3: responding to these surveys. And one thing I always wondered 162 00:09:37,559 --> 00:09:42,880 Speaker 3: about is how how I guess onerous or resource intensive 163 00:09:43,480 --> 00:09:46,040 Speaker 3: is responding to these surveys. So do I have to 164 00:09:46,080 --> 00:09:50,079 Speaker 3: have like a person who's dedicated to doing this every month? 165 00:09:50,280 --> 00:09:52,839 Speaker 3: Does it consume resources? How long is this going to 166 00:09:52,880 --> 00:09:56,280 Speaker 3: take me? And then also does the BLS ever try 167 00:09:56,320 --> 00:10:00,760 Speaker 3: to fact check some of these self reported data points? 168 00:10:02,320 --> 00:10:06,400 Speaker 5: Those are great questions. So the survey is not just 169 00:10:06,559 --> 00:10:09,760 Speaker 5: one or two questions. It takes about thirty minutes to 170 00:10:09,800 --> 00:10:13,960 Speaker 5: fill it out completely. We do get responses that are incomplete, 171 00:10:13,960 --> 00:10:18,160 Speaker 5: and we accept those incomplete responses. The survey is usually 172 00:10:18,240 --> 00:10:22,880 Speaker 5: filled out by someone who can access the company's data 173 00:10:23,480 --> 00:10:29,040 Speaker 5: on wages, on total employment, on employment by supervisory or 174 00:10:29,240 --> 00:10:33,120 Speaker 5: non supervisory job functions. In other words, it's a person 175 00:10:33,360 --> 00:10:37,640 Speaker 5: usually in the operations office of a company. Now we 176 00:10:37,679 --> 00:10:39,880 Speaker 5: ask companies that are very large, you know, like the 177 00:10:39,960 --> 00:10:42,960 Speaker 5: big box retailers, all the way down to really small 178 00:10:43,040 --> 00:10:47,719 Speaker 5: corporations or small firms, and it becomes more and more 179 00:10:47,760 --> 00:10:50,320 Speaker 5: of a difficulty for the smaller firms, and we're very 180 00:10:50,360 --> 00:10:54,240 Speaker 5: cognizant of that. So BLS has reduced the scope of 181 00:10:54,240 --> 00:10:57,679 Speaker 5: the survey to get it as small as possible and 182 00:10:57,760 --> 00:11:02,320 Speaker 5: as free of burdens and compliance as possible. It's also 183 00:11:02,360 --> 00:11:05,400 Speaker 5: an electronics survey so that they can fill it out 184 00:11:05,440 --> 00:11:09,160 Speaker 5: on their computer screens and hit a submit button, which 185 00:11:09,240 --> 00:11:13,080 Speaker 5: greatly greatly improved our response rate, by the way, so 186 00:11:13,200 --> 00:11:15,640 Speaker 5: that survey is not so hard now to fill out, 187 00:11:15,640 --> 00:11:18,600 Speaker 5: though I think it still takes about thirty minutes, maybe 188 00:11:18,640 --> 00:11:20,360 Speaker 5: more if you have to dig out the data if 189 00:11:20,360 --> 00:11:23,560 Speaker 5: you're not in possession of it. The household survey is 190 00:11:23,640 --> 00:11:27,360 Speaker 5: a totally different beast. It is a long survey. It 191 00:11:27,400 --> 00:11:31,040 Speaker 5: consists of over one hundred and twenty questions. It really 192 00:11:31,600 --> 00:11:35,480 Speaker 5: is hard to fill out if you kind of don't 193 00:11:35,480 --> 00:11:38,559 Speaker 5: know the terminology. Let me just give you one. We 194 00:11:38,600 --> 00:11:41,960 Speaker 5: asked the question on employment. Are you working or have 195 00:11:42,080 --> 00:11:44,319 Speaker 5: you looked for work in the past four weeks? That's 196 00:11:44,400 --> 00:11:46,720 Speaker 5: the key question. We've been asking that question for since 197 00:11:46,800 --> 00:11:51,000 Speaker 5: nineteen I think nineteen forty three. If they answer I'm working, 198 00:11:51,280 --> 00:11:54,640 Speaker 5: then they are considered employed. We'll have follow up questions 199 00:11:54,679 --> 00:11:57,839 Speaker 5: about that, about their industry, about their tasks and so forth. 200 00:11:58,559 --> 00:12:01,880 Speaker 5: If they say no, I'm not working, then we say, 201 00:12:01,880 --> 00:12:04,120 Speaker 5: well did you look for work in the past four weeks? 202 00:12:04,520 --> 00:12:09,439 Speaker 5: And they'll say, well, you know, maybe I thought about it. 203 00:12:09,640 --> 00:12:12,600 Speaker 5: Well thinking about it. It's not enough you can look 204 00:12:12,640 --> 00:12:14,760 Speaker 5: for one day in the past four weeks and be 205 00:12:15,440 --> 00:12:19,720 Speaker 5: considered in the labor force but unemployed. So that sometimes 206 00:12:19,720 --> 00:12:23,120 Speaker 5: we have to have a verbal follow up rather than 207 00:12:23,160 --> 00:12:26,200 Speaker 5: they're just you know, just filling out a form. There 208 00:12:26,240 --> 00:12:29,160 Speaker 5: has to be a conversation. That's why we don't do 209 00:12:29,240 --> 00:12:32,239 Speaker 5: more than sixty thousand households because it is it's burdensome 210 00:12:32,880 --> 00:12:35,400 Speaker 5: for the person who's the questioner. 211 00:12:35,640 --> 00:12:38,480 Speaker 2: This leads right into I think what I was about 212 00:12:38,520 --> 00:12:40,679 Speaker 2: to ask you, which is, you know we're having this 213 00:12:40,760 --> 00:12:44,640 Speaker 2: conversation April twenty twenty five. Your term of office at 214 00:12:44,640 --> 00:12:47,400 Speaker 2: the BLS ended March twenty twenty three, so you're actually 215 00:12:47,440 --> 00:12:50,360 Speaker 2: under both the two recent presidents, under the first Trump 216 00:12:50,400 --> 00:12:54,120 Speaker 2: administration then Biden. But let's say, okay, so let's say 217 00:12:54,120 --> 00:12:58,640 Speaker 2: it's March twenty twenty three. How much costlier is that 218 00:12:58,800 --> 00:13:02,680 Speaker 2: process of commune unicating with the households than it was, 219 00:13:02,800 --> 00:13:05,680 Speaker 2: say in twenty thirteen or two thousand and three. 220 00:13:07,720 --> 00:13:10,720 Speaker 5: It was more and more costly, certainly, and that's because 221 00:13:10,760 --> 00:13:14,960 Speaker 5: of inflation and how inflation affected the wage bill for 222 00:13:15,040 --> 00:13:19,400 Speaker 5: the survey. Okay, the survey is taken by the Census 223 00:13:19,480 --> 00:13:23,240 Speaker 5: Bureau under a contract by BLS, So we pay the 224 00:13:23,240 --> 00:13:25,760 Speaker 5: Census Bureau to take this to go out in the 225 00:13:25,760 --> 00:13:29,480 Speaker 5: field and conduct the survey. Here's an idea of how 226 00:13:29,520 --> 00:13:32,480 Speaker 5: much more expensive it was. In twenty nineteen, which is 227 00:13:32,480 --> 00:13:34,800 Speaker 5: the first year of my term, I didn't have to 228 00:13:34,840 --> 00:13:39,760 Speaker 5: find any additional funding for the household survey. It might 229 00:13:39,760 --> 00:13:44,040 Speaker 5: have been just a small amount, but nothing significant. By 230 00:13:44,160 --> 00:13:46,640 Speaker 5: twenty twenty three, we were having to find almost four 231 00:13:46,679 --> 00:13:50,280 Speaker 5: million five million dollars. More so, the costs of the 232 00:13:50,320 --> 00:13:53,760 Speaker 5: survey is that ye per month, that would be for 233 00:13:53,840 --> 00:13:58,319 Speaker 5: the entire year. The cost of the survey went up significantly. 234 00:13:58,440 --> 00:14:00,840 Speaker 5: Now your listeners will be thinking about the federal budget, 235 00:14:00,840 --> 00:14:03,239 Speaker 5: which is in the trillions of dollars. You'd say, oh, 236 00:14:03,280 --> 00:14:07,000 Speaker 5: that's not very much money. But the Bureau's budget had 237 00:14:07,400 --> 00:14:12,600 Speaker 5: has not really changed for almost a decade, and our 238 00:14:12,640 --> 00:14:15,880 Speaker 5: costs have gone up, you know, progressively, especially during the 239 00:14:15,920 --> 00:14:18,839 Speaker 5: period of significant inflation which we had starting in twenty 240 00:14:18,880 --> 00:14:22,360 Speaker 5: twenty one, so that you know, finding additional money is 241 00:14:22,440 --> 00:14:26,720 Speaker 5: really hard. During the pandemic, I found millions of dollars 242 00:14:26,720 --> 00:14:31,680 Speaker 5: of couch money, unused conference fees, unused travel fees, so 243 00:14:31,920 --> 00:14:35,960 Speaker 5: I could apply that couch money if it were, you know, 244 00:14:36,040 --> 00:14:40,240 Speaker 5: to p purposes like buoying up the CPS or building 245 00:14:40,280 --> 00:14:42,960 Speaker 5: a data We built a brand new data center out 246 00:14:43,000 --> 00:14:48,760 Speaker 5: in the suburb of Washington. But once we were back 247 00:14:48,800 --> 00:14:53,120 Speaker 5: in business and fully using our entire budget, it was 248 00:14:53,240 --> 00:14:56,760 Speaker 5: very difficult to find those dollars to fill the hole. 249 00:14:57,320 --> 00:14:59,880 Speaker 5: And I think you could safely say I've been arguing 250 00:14:59,880 --> 00:15:03,320 Speaker 5: this now for over two years. Our surveys, but especially 251 00:15:03,720 --> 00:15:06,200 Speaker 5: the current population survey, the one that gives us the 252 00:15:06,280 --> 00:15:10,920 Speaker 5: unemployment rate, the labor force data are dying, They're decaying. 253 00:15:11,280 --> 00:15:14,200 Speaker 5: They're in very serious trouble and we will have if 254 00:15:14,600 --> 00:15:17,200 Speaker 5: unless we modernize that survey, we will see a time 255 00:15:17,240 --> 00:15:20,480 Speaker 5: when we will be like the British right, unable to 256 00:15:20,600 --> 00:15:24,720 Speaker 5: publish portions of it that just don't have sufficient sample 257 00:15:25,400 --> 00:15:27,120 Speaker 5: for statistical release. 258 00:15:27,760 --> 00:15:30,520 Speaker 3: Okay, I'm going to ask the obvious question here, but 259 00:15:30,760 --> 00:15:34,320 Speaker 3: why have costs of conducting these surveys gathering this data 260 00:15:34,360 --> 00:15:37,840 Speaker 3: actually gone up? And I understand, you know, building data 261 00:15:37,880 --> 00:15:41,720 Speaker 3: centers is probably an expensive process. But on the other hand, 262 00:15:41,960 --> 00:15:45,760 Speaker 3: you know, going electronic instead of mailing out thousands and 263 00:15:45,840 --> 00:15:49,080 Speaker 3: thousands of surveys in theory, I would imagine maybe that 264 00:15:49,240 --> 00:15:53,280 Speaker 3: saves you some money. So what exactly is causing the 265 00:15:53,440 --> 00:15:54,920 Speaker 3: increase in expenses here? 266 00:15:56,360 --> 00:15:59,440 Speaker 5: The problem is not with the electronic surveys. They will 267 00:15:59,480 --> 00:16:03,960 Speaker 5: eventually be very costly and they're there. We're suffering from 268 00:16:04,000 --> 00:16:06,440 Speaker 5: a different problem, which is just public support for the 269 00:16:06,480 --> 00:16:11,320 Speaker 5: electronic surveys. So the jobs report, that is that portion 270 00:16:11,400 --> 00:16:14,920 Speaker 5: of it that's electronically captured by these collection centers in 271 00:16:15,080 --> 00:16:18,680 Speaker 5: Chicago and Atlanta. I think those are going along all right, 272 00:16:18,920 --> 00:16:24,280 Speaker 5: for the time being, Our real problem is in the surveys, 273 00:16:24,760 --> 00:16:28,760 Speaker 5: the data of which is collected by people survey survey, 274 00:16:28,840 --> 00:16:32,880 Speaker 5: field survey people. They go out and they talk to 275 00:16:32,920 --> 00:16:36,480 Speaker 5: the households and they're The cost is the obvious one. 276 00:16:36,520 --> 00:16:42,200 Speaker 5: It is the wage bill. These people are highly skilled interviewers. 277 00:16:42,320 --> 00:16:44,840 Speaker 5: They are not high school graduates, and I have nothing 278 00:16:44,880 --> 00:16:48,440 Speaker 5: against high school graduates, but they are educated, very trained, 279 00:16:48,920 --> 00:16:53,960 Speaker 5: oftentimes economists, who will go out and speak to people 280 00:16:54,400 --> 00:16:58,840 Speaker 5: and to follow up interviews, oh my gosh, and they'll 281 00:16:58,880 --> 00:17:01,880 Speaker 5: spend hours and hours do it this. Well, okay, fine, 282 00:17:01,920 --> 00:17:05,520 Speaker 5: that's great, but it costs a lot to keep those 283 00:17:05,560 --> 00:17:09,080 Speaker 5: people employed. They have other opportunities, so we have to 284 00:17:09,160 --> 00:17:13,520 Speaker 5: have competitive wages. The wage bill is driving that. Collection 285 00:17:13,680 --> 00:17:17,080 Speaker 5: costs are a little bit higher, that is the processing costs, 286 00:17:17,520 --> 00:17:19,640 Speaker 5: but the big part of that is the wage build 287 00:17:19,640 --> 00:17:21,240 Speaker 5: and the wage bill is not going to go away. 288 00:17:21,400 --> 00:17:24,960 Speaker 5: That's just going to continue to go up. So we 289 00:17:25,000 --> 00:17:28,240 Speaker 5: need to do with the household survey what we have 290 00:17:28,359 --> 00:17:32,600 Speaker 5: done with the establishment survey, the firm survey, and that 291 00:17:32,760 --> 00:17:37,439 Speaker 5: is modernize it so it is more electronically collected. And 292 00:17:37,480 --> 00:17:40,960 Speaker 5: then we also need to integrate data which can be 293 00:17:41,000 --> 00:17:44,800 Speaker 5: obtained through the Internet on households. Households maybe go on 294 00:17:44,880 --> 00:17:48,040 Speaker 5: to a platform and with a tablet or something and 295 00:17:48,160 --> 00:17:50,959 Speaker 5: supply Maybe we go to a million households and they 296 00:17:51,000 --> 00:17:54,280 Speaker 5: supply five or six questions and we combine that or 297 00:17:54,320 --> 00:18:00,360 Speaker 5: blended with a person to person collected data. That's that's 298 00:18:00,359 --> 00:18:04,080 Speaker 5: what we call modernization. I mean, I mean that sounds 299 00:18:04,080 --> 00:18:08,080 Speaker 5: so innovative to your listeners, but it is very innovative 300 00:18:08,080 --> 00:18:11,480 Speaker 5: for the statistical system. Unless we do that, the cost 301 00:18:11,560 --> 00:18:14,000 Speaker 5: will continue to rise. The response rates, by the way, 302 00:18:14,080 --> 00:18:17,919 Speaker 5: are continuing to decline. So that's public support and we 303 00:18:18,000 --> 00:18:21,560 Speaker 5: will be we just won't have these surveys in the future. 304 00:18:21,600 --> 00:18:22,720 Speaker 5: There is too expensive. 305 00:18:22,840 --> 00:18:25,159 Speaker 2: I want to get to the modernization in a second, 306 00:18:25,240 --> 00:18:28,320 Speaker 2: but just talk to us about I mean, I think 307 00:18:28,320 --> 00:18:31,280 Speaker 2: you said these the surveys dying, which is pretty dramatic. 308 00:18:31,400 --> 00:18:36,320 Speaker 2: Given the centrality of this between the response rates, the 309 00:18:36,480 --> 00:18:41,320 Speaker 2: increased wage bill, et cetera. How severe is this crisis? 310 00:18:41,359 --> 00:18:43,600 Speaker 2: How much time are we talking about because I imagine 311 00:18:43,600 --> 00:18:46,200 Speaker 2: it gets harder and harder to find that couch change 312 00:18:46,240 --> 00:18:46,880 Speaker 2: to keep it going. 313 00:18:48,119 --> 00:18:50,920 Speaker 5: Yeah, Actually, we're out of time. We have to cut 314 00:18:50,960 --> 00:18:54,200 Speaker 5: down on the sample and I'll give you I'll give 315 00:18:54,240 --> 00:19:00,320 Speaker 5: you the evidence, my successor, the current Commissioner of Labor Statistics, 316 00:19:00,359 --> 00:19:03,480 Speaker 5: at the end of last year that we would have 317 00:19:03,560 --> 00:19:07,720 Speaker 5: to reduce the sample in the CPS, that's the household 318 00:19:07,800 --> 00:19:11,359 Speaker 5: survey by five thousand households in order to just publish 319 00:19:11,800 --> 00:19:16,199 Speaker 5: the rest of the sample. Okay, by cutting back on 320 00:19:16,280 --> 00:19:22,080 Speaker 5: the sample, you cut back on publishing details in the 321 00:19:22,160 --> 00:19:26,600 Speaker 5: current report. You may not have enough to publish on 322 00:19:26,680 --> 00:19:31,919 Speaker 5: teenagers anymore. On certain demographic groups, you may not be 323 00:19:32,000 --> 00:19:36,360 Speaker 5: able to publish details on certain age intervals, like everyone 324 00:19:36,480 --> 00:19:39,840 Speaker 5: between the ages of fifty five and sixty or sixty 325 00:19:39,880 --> 00:19:44,120 Speaker 5: five and seventy five. Every time you reduce the sample 326 00:19:44,560 --> 00:19:47,800 Speaker 5: or your response rate falls and it kind of reduces 327 00:19:47,840 --> 00:19:52,360 Speaker 5: it for you. We can talk about that next. What 328 00:19:52,440 --> 00:19:55,600 Speaker 5: suffers are the details. Let's go to the CPI, which 329 00:19:55,640 --> 00:20:00,000 Speaker 5: BLS also does. The CPI is a very big survey, 330 00:20:00,080 --> 00:20:03,240 Speaker 5: important survey, but it is oftentimes the case that we 331 00:20:03,359 --> 00:20:07,280 Speaker 5: don't have details for certain parts of the basket of goods, 332 00:20:08,080 --> 00:20:12,200 Speaker 5: and so we will average across some months and published 333 00:20:12,240 --> 00:20:15,760 Speaker 5: for another month based on averages of prior months. But 334 00:20:15,800 --> 00:20:18,960 Speaker 5: then we run out of the integrity of that average 335 00:20:19,320 --> 00:20:23,440 Speaker 5: and we can't publish the bread price or something like that. 336 00:20:23,440 --> 00:20:28,199 Speaker 5: That happens more and more because we're getting less and 337 00:20:28,280 --> 00:20:33,600 Speaker 5: less cooperation by retailers to allow our surveyors into their 338 00:20:33,640 --> 00:20:36,800 Speaker 5: stores and to go around grabbing the potato chip bag 339 00:20:36,840 --> 00:20:39,200 Speaker 5: and counting the number of potatoes in the back. So 340 00:20:40,520 --> 00:20:44,400 Speaker 5: we've had to innovate on the CPI, on the producer 341 00:20:44,480 --> 00:20:48,920 Speaker 5: price Index, on the import export price indexes. I hope 342 00:20:48,960 --> 00:20:51,679 Speaker 5: you haven't noticed the innovations because you're not supposed to, 343 00:20:52,320 --> 00:20:57,159 Speaker 5: but those innovations have saved, absolutely saved the import and 344 00:20:57,200 --> 00:21:00,679 Speaker 5: export price. We had response rates down in the twenty centiles, 345 00:21:01,200 --> 00:21:05,399 Speaker 5: completely unpublishable detail. During the time I was commissioner. We 346 00:21:05,440 --> 00:21:09,439 Speaker 5: went to using data from the Commerce Department that is 347 00:21:09,480 --> 00:21:13,119 Speaker 5: collected by the customs officials and it's working out just great, 348 00:21:13,560 --> 00:21:17,760 Speaker 5: so we completely left the surveys there. The PPI has 349 00:21:17,840 --> 00:21:21,080 Speaker 5: problems just like the British are having. Now not as bad, 350 00:21:21,600 --> 00:21:24,960 Speaker 5: and we're using more and more data from private companies 351 00:21:25,720 --> 00:21:29,080 Speaker 5: who will give us their data from the Internet, where 352 00:21:29,119 --> 00:21:31,760 Speaker 5: we collect that data and we combine that with the 353 00:21:31,800 --> 00:21:36,120 Speaker 5: survey data to keep the things alive on the CPI. 354 00:21:36,520 --> 00:21:40,200 Speaker 5: It's very interesting. We get all of our retail gas 355 00:21:40,240 --> 00:21:44,680 Speaker 5: prices now from a private company that aggregates gas prices 356 00:21:44,760 --> 00:21:49,000 Speaker 5: for the retail gasoline industry, and it's a very good source. 357 00:21:49,160 --> 00:21:51,600 Speaker 5: A lot of our housing prices now you mentioned housing, 358 00:21:52,000 --> 00:21:56,080 Speaker 5: We're going to more and more to aggregators, private aggregators. 359 00:21:56,119 --> 00:21:59,400 Speaker 5: So there are strategies for saving the survey by blending 360 00:21:59,440 --> 00:22:02,840 Speaker 5: in private data with the survey data. That is not 361 00:22:03,000 --> 00:22:06,440 Speaker 5: the case with the household survey of the labor force. 362 00:22:07,359 --> 00:22:11,680 Speaker 5: Nobody really collects that data except BLS. And that's why 363 00:22:11,760 --> 00:22:16,280 Speaker 5: these demographic surveys, Census has a bunch of them. Health 364 00:22:16,640 --> 00:22:20,680 Speaker 5: the health department, the Department of Health has a lot 365 00:22:21,240 --> 00:22:22,639 Speaker 5: very in very serious trouble. 366 00:22:23,680 --> 00:22:26,760 Speaker 3: So I take the point about innovations. But one of 367 00:22:26,840 --> 00:22:29,560 Speaker 3: the things that's been happening recently is we have been 368 00:22:29,600 --> 00:22:33,720 Speaker 3: seeing bigger and bigger revisions in a lot of the data. 369 00:22:33,880 --> 00:22:37,520 Speaker 3: Is that because of the response issues, So more late 370 00:22:37,760 --> 00:22:41,159 Speaker 3: responses mean that you know you're going to get revisions 371 00:22:41,240 --> 00:22:45,040 Speaker 3: later on as those responses are incorporated. And the reason 372 00:22:45,080 --> 00:22:48,960 Speaker 3: I ask is I remember speaking to someone at the BLS, 373 00:22:49,040 --> 00:22:52,000 Speaker 3: and here shout out to the people working at the BLS, 374 00:22:52,040 --> 00:22:55,040 Speaker 3: because you can actually just call them and ask questions. 375 00:22:55,040 --> 00:22:59,520 Speaker 3: They're really really responsive, amazingly, like the only government department 376 00:22:59,520 --> 00:23:01,439 Speaker 3: that I really know where they'll just pick up the 377 00:23:01,440 --> 00:23:04,760 Speaker 3: phone and talk to you about methodology. Anyway, I was 378 00:23:04,800 --> 00:23:08,880 Speaker 3: asking about revisions to PPI for mayonnaise, of all things, 379 00:23:09,280 --> 00:23:11,879 Speaker 3: and the data there had been revised from like five 380 00:23:11,960 --> 00:23:16,080 Speaker 3: percent to ten percent, which is a pretty big change 381 00:23:16,320 --> 00:23:18,840 Speaker 3: in the space of a couple months. And he was 382 00:23:18,880 --> 00:23:21,680 Speaker 3: talking about how the PPI is subject to revision up 383 00:23:21,680 --> 00:23:25,400 Speaker 3: to four months after it's published, and it gets updated 384 00:23:25,400 --> 00:23:28,520 Speaker 3: as new replies come in. So I imagine if response 385 00:23:28,600 --> 00:23:31,879 Speaker 3: rates are going lower, then maybe that's contributing to some 386 00:23:32,000 --> 00:23:33,320 Speaker 3: of the revision issue. 387 00:23:33,560 --> 00:23:36,800 Speaker 5: By and large, our issues have to do with response rate. 388 00:23:37,480 --> 00:23:40,480 Speaker 5: You're absolutely right, and I'm before I go further, I'm 389 00:23:40,560 --> 00:23:44,120 Speaker 5: very very happy you said that about BLS because BLS 390 00:23:44,160 --> 00:23:47,760 Speaker 5: prize itself and being responsive and transparent by the way, 391 00:23:47,880 --> 00:23:50,399 Speaker 5: so that's a good thing. So I think there are 392 00:23:50,400 --> 00:23:54,439 Speaker 5: two sources of this problem. The first is really hidden 393 00:23:54,480 --> 00:23:56,240 Speaker 5: and no one talks about it. So let me just 394 00:23:56,320 --> 00:24:01,439 Speaker 5: mention when we refresh survey, and you have to do 395 00:24:01,480 --> 00:24:04,600 Speaker 5: this because people are asked to give their survey responses 396 00:24:04,640 --> 00:24:06,760 Speaker 5: only for a few months, right, and then they drop 397 00:24:06,800 --> 00:24:10,639 Speaker 5: out and you have to find new people. The regional offices, 398 00:24:10,680 --> 00:24:12,600 Speaker 5: there are six of them in the United States for 399 00:24:12,640 --> 00:24:15,639 Speaker 5: the BLS, they have people who go out and do 400 00:24:15,760 --> 00:24:20,400 Speaker 5: what's called initiate a new survey respondent. It's getting more 401 00:24:20,440 --> 00:24:24,160 Speaker 5: and more difficult now to initiate that respondent. Well, people 402 00:24:24,200 --> 00:24:26,200 Speaker 5: are saying, we just don't have time, you know, Oh 403 00:24:26,200 --> 00:24:28,600 Speaker 5: my gosh, there's so many important things to do. We 404 00:24:28,640 --> 00:24:30,679 Speaker 5: don't want to give you our data because if we 405 00:24:30,720 --> 00:24:32,680 Speaker 5: give you our data, then the IRS is going to 406 00:24:32,720 --> 00:24:34,320 Speaker 5: be after us, or we're going to have some kind 407 00:24:34,320 --> 00:24:38,000 Speaker 5: of osha of you know, we don't. BLS's data is 408 00:24:38,040 --> 00:24:43,000 Speaker 5: completely protected from the law enforcement aspects of the federal government, 409 00:24:43,440 --> 00:24:47,240 Speaker 5: absolutely only for statistical purposes. That does not prevent help 410 00:24:47,760 --> 00:24:49,560 Speaker 5: or people saying, oh, it's going to happen, you know. 411 00:24:50,200 --> 00:24:54,240 Speaker 5: So if your initiations are dropping and it's harder and harder, 412 00:24:54,560 --> 00:24:57,680 Speaker 5: no wonder the response rates are dropping because people are 413 00:24:57,760 --> 00:25:00,440 Speaker 5: less enthusiastic. Even if they say yeah, I'd love to 414 00:25:00,480 --> 00:25:03,760 Speaker 5: be a part of it, they're not as enthusiastic as 415 00:25:03,800 --> 00:25:07,320 Speaker 5: they were a generation ago. So the real problem starts 416 00:25:07,359 --> 00:25:10,720 Speaker 5: at the initiation level, and we're seeing that not only 417 00:25:10,800 --> 00:25:15,240 Speaker 5: in the household survey and in the price area, but 418 00:25:15,280 --> 00:25:19,080 Speaker 5: we're also seeing that in another survey we haven't talked about, 419 00:25:19,080 --> 00:25:22,560 Speaker 5: the National Compensation Survey, which is a wonderful survey on 420 00:25:22,720 --> 00:25:27,120 Speaker 5: how much people are getting in union houses, union households 421 00:25:27,160 --> 00:25:30,480 Speaker 5: and non union households, and Northeast and the Southwest is 422 00:25:30,600 --> 00:25:35,520 Speaker 5: terrific about that. The Joelts report the job opening is 423 00:25:35,520 --> 00:25:37,919 Speaker 5: in labor turnover survey that you mentioned at the opening. 424 00:25:38,480 --> 00:25:42,960 Speaker 5: Its response rate has fallen dramatically, and it's largely because 425 00:25:43,000 --> 00:25:49,000 Speaker 5: people are less enthusiastic generally about participating in government surveys. 426 00:25:49,280 --> 00:25:52,399 Speaker 5: It's going to be hard to stem that tide, and 427 00:25:52,440 --> 00:25:54,679 Speaker 5: I think the only way we can do it is 428 00:25:54,720 --> 00:25:59,840 Speaker 5: by going to these platformed surveys, that is, surveys that 429 00:26:00,119 --> 00:26:03,560 Speaker 5: use the Internet as the main platform, integrating a lot 430 00:26:03,560 --> 00:26:07,600 Speaker 5: of private data, and being highly original in the way 431 00:26:07,640 --> 00:26:12,280 Speaker 5: we think about preparing a statistic for public use. We're 432 00:26:12,320 --> 00:26:14,879 Speaker 5: just not going to be facing a different world overnight 433 00:26:15,000 --> 00:26:17,560 Speaker 5: of really happy people wanting to give all their data 434 00:26:17,600 --> 00:26:18,159 Speaker 5: to the government. 435 00:26:18,560 --> 00:26:21,840 Speaker 2: It's really a fallen world. Sometimes I'll say that, you know, 436 00:26:22,200 --> 00:26:24,040 Speaker 2: we won't let you in the store to you know, 437 00:26:24,080 --> 00:26:26,439 Speaker 2: look at the price of bread anyway? What would it 438 00:26:26,480 --> 00:26:29,600 Speaker 2: take to get there? Can this be done unilaterally within 439 00:26:29,640 --> 00:26:32,480 Speaker 2: the BLS, this sort of big upgrade. Would it need 440 00:26:32,520 --> 00:26:34,600 Speaker 2: to be something involved with Congress? Would it need a 441 00:26:34,640 --> 00:26:38,560 Speaker 2: budget allocation that would come from Congress? What would need 442 00:26:38,600 --> 00:26:42,640 Speaker 2: to happen to get the sort of modernized statistical collection 443 00:26:42,720 --> 00:26:44,720 Speaker 2: system that you would like actually in place? 444 00:26:44,960 --> 00:26:47,040 Speaker 5: I think we have to approach this like we might 445 00:26:47,160 --> 00:26:52,360 Speaker 5: approach roads and bridges right. Once Congress becomes aware that 446 00:26:52,720 --> 00:26:57,000 Speaker 5: there's political liability and ignoring a problem, they generally focus 447 00:26:57,040 --> 00:26:59,520 Speaker 5: on it until it's fixed. And that was the case 448 00:26:59,560 --> 00:27:01,720 Speaker 5: with our national highway system still is a case with 449 00:27:01,800 --> 00:27:04,240 Speaker 5: our national highway system. We still have a lot of problems. 450 00:27:04,640 --> 00:27:06,600 Speaker 5: When I go to Congress and I talk about the 451 00:27:06,640 --> 00:27:08,879 Speaker 5: CPS and response rate, how we're going to lose the 452 00:27:08,920 --> 00:27:14,560 Speaker 5: unemployment rate, I get immediate response, and nobody, the Communists, 453 00:27:14,680 --> 00:27:18,359 Speaker 5: no one's ever said that to me, Oh, let's fix it. 454 00:27:18,760 --> 00:27:23,280 Speaker 5: So in the last Congressional Continuing Resolution, which is passed 455 00:27:23,280 --> 00:27:26,760 Speaker 5: a few weeks ago, BLS got six million more in 456 00:27:26,960 --> 00:27:29,760 Speaker 5: funding just to fill that hole that we've been talking 457 00:27:29,800 --> 00:27:33,840 Speaker 5: about on the wage bill. And that was an amazing 458 00:27:33,880 --> 00:27:37,400 Speaker 5: thing to get in a continuing resolution and increase in funding. 459 00:27:37,880 --> 00:27:42,080 Speaker 5: So I think Congress, presented with a plan, or the 460 00:27:42,119 --> 00:27:47,120 Speaker 5: administration of President Trump, is wide open to disruption and change. 461 00:27:47,640 --> 00:27:51,200 Speaker 5: I think if we develop an aggressive, bold, comprehensive plan 462 00:27:51,280 --> 00:27:54,879 Speaker 5: about how to rebuild the statistical system so that we're 463 00:27:55,600 --> 00:27:59,560 Speaker 5: using our resources much more efficiently, perhaps combining some agencies 464 00:27:59,600 --> 00:28:03,840 Speaker 5: together instead of having the twenty four separate statistical agencies, 465 00:28:03,840 --> 00:28:06,640 Speaker 5: maybe we ought to have just a handful, and then 466 00:28:06,800 --> 00:28:10,280 Speaker 5: going from there to a highly innovative different way of 467 00:28:10,359 --> 00:28:15,080 Speaker 5: collecting and disseminating data. Then our roads and bridges, statistically speaking, 468 00:28:15,119 --> 00:28:19,160 Speaker 5: won't be disintegrating or decaying. We will have new concrete, 469 00:28:19,160 --> 00:28:22,159 Speaker 5: will have new structures, and you can see a future 470 00:28:22,200 --> 00:28:24,880 Speaker 5: for the statistical system. I think right now, speaking as 471 00:28:24,920 --> 00:28:27,960 Speaker 5: a former director of an agency one of the most 472 00:28:27,960 --> 00:28:31,400 Speaker 5: important statistical agencies and not the most important statistical agency 473 00:28:31,400 --> 00:28:34,920 Speaker 5: in the world, that future looks grim to me, and 474 00:28:35,280 --> 00:28:38,000 Speaker 5: so change is required. It has to happen, and I 475 00:28:38,040 --> 00:28:40,280 Speaker 5: think that's what we have to do. Presented with the plan, 476 00:28:40,720 --> 00:28:42,920 Speaker 5: let Congress see what we're going to do and have 477 00:28:43,040 --> 00:28:46,320 Speaker 5: them fund modernization, not continuation of the current system. 478 00:28:47,080 --> 00:28:49,520 Speaker 3: This is a little out there, but could you if 479 00:28:49,560 --> 00:28:52,840 Speaker 3: the problem is the response rates and incentivizing people to 480 00:28:53,080 --> 00:28:57,640 Speaker 3: actually answer these surveys, could you potentially pay them to 481 00:28:57,720 --> 00:28:59,560 Speaker 3: do it. 482 00:29:00,080 --> 00:29:05,560 Speaker 5: We've tried that. It's been tried a lot. Good suggestion. Yeah, 483 00:29:05,640 --> 00:29:09,200 Speaker 5: you know, pay them dollars. Okay, tried that, didn't make 484 00:29:09,480 --> 00:29:11,480 Speaker 5: much of a difference. Then we thought, well, maybe they'd 485 00:29:11,520 --> 00:29:16,240 Speaker 5: like to have these cards you can take to retailers 486 00:29:16,240 --> 00:29:19,560 Speaker 5: and buy anything you want, right, gift cards. We tried, Yeah, 487 00:29:19,640 --> 00:29:24,280 Speaker 5: gift cards. So we tried that. Now that's not the issue. 488 00:29:25,640 --> 00:29:27,880 Speaker 5: I think you could pay people a lot of money, 489 00:29:27,960 --> 00:29:30,520 Speaker 5: say one thousand dollars a month, and they might participate 490 00:29:30,520 --> 00:29:33,400 Speaker 5: because that's a lot of money, But we can't afford that. 491 00:29:33,120 --> 00:29:37,680 Speaker 5: That's not out there for the statistical system. So inducements 492 00:29:38,600 --> 00:29:41,840 Speaker 5: may help with the margin, but they don't change the 493 00:29:41,960 --> 00:29:45,440 Speaker 5: trend line, which is going negative on the response rate. 494 00:29:46,080 --> 00:29:47,920 Speaker 5: I think we're going to live at that response rate 495 00:29:48,320 --> 00:29:51,560 Speaker 5: for a while. I do believe it's generational. I think 496 00:29:51,880 --> 00:29:54,600 Speaker 5: you can see in the really young kids now, not 497 00:29:55,040 --> 00:29:57,280 Speaker 5: the ones that are under five, but the ones that 498 00:29:57,320 --> 00:30:00,560 Speaker 5: are in their teams, kind of a return to doing 499 00:30:00,640 --> 00:30:04,680 Speaker 5: things together, having more social events. Maybe I would say 500 00:30:05,200 --> 00:30:08,080 Speaker 5: the bowling leagues are coming back, and maybe that's a 501 00:30:08,080 --> 00:30:11,040 Speaker 5: cultural change that leads to a more a greater sense 502 00:30:11,040 --> 00:30:15,000 Speaker 5: of participation and support of public institutions, one could hope. 503 00:30:15,200 --> 00:30:18,280 Speaker 3: I like the idea of all the kids coming together 504 00:30:18,600 --> 00:30:21,960 Speaker 3: to answer surveys from the BLS. That's great. 505 00:30:22,920 --> 00:30:25,720 Speaker 5: I think you get some really interesting answers there. But yes, 506 00:30:26,800 --> 00:30:29,520 Speaker 5: short of that, we have to be innovative. We have 507 00:30:29,560 --> 00:30:32,880 Speaker 5: to change, We have to think outside the box. Otherwise 508 00:30:33,120 --> 00:30:36,600 Speaker 5: this infrastructure which we all need, you know, it runs 509 00:30:36,600 --> 00:30:39,960 Speaker 5: our country. There's no economy without the Bureau of Economic 510 00:30:40,040 --> 00:30:43,400 Speaker 5: Analysis and the GDP numbers. They don't grow on trees. 511 00:30:44,040 --> 00:30:46,640 Speaker 5: That's going to be either going to go away or 512 00:30:46,680 --> 00:30:47,840 Speaker 5: become less reliable. 513 00:31:04,000 --> 00:31:07,400 Speaker 2: You know something that strikes me while listening to you talk, 514 00:31:07,560 --> 00:31:11,280 Speaker 2: and like, oh, you found a way to allocate you know, 515 00:31:11,400 --> 00:31:13,960 Speaker 2: unused travel spending so that you could keep the surveys 516 00:31:14,000 --> 00:31:16,280 Speaker 2: going after the wage bilt went up, et cetera. I 517 00:31:16,360 --> 00:31:19,760 Speaker 2: keep rustling the couch change and I think about this 518 00:31:20,000 --> 00:31:22,080 Speaker 2: in you know, you mentioned Okay, well maybe the new 519 00:31:22,160 --> 00:31:26,880 Speaker 2: president is you know, theoretically open to shaking things up 520 00:31:26,920 --> 00:31:29,680 Speaker 2: and doing things a different way. Would you argue, pretty persuasively, 521 00:31:29,760 --> 00:31:32,120 Speaker 2: isn't necessary on the other hand, you have this sort 522 00:31:32,160 --> 00:31:34,360 Speaker 2: of dose kick which is sort of premised on this 523 00:31:34,520 --> 00:31:39,000 Speaker 2: idea that every agency in the government somehow is just 524 00:31:39,240 --> 00:31:43,400 Speaker 2: egregiously wasteful. That all of these agencies must be so 525 00:31:43,560 --> 00:31:47,120 Speaker 2: wasteful that you could cut aggressively and you almost certainly 526 00:31:47,160 --> 00:31:49,959 Speaker 2: aren't actually going to hit any bone. Seems to be 527 00:31:50,280 --> 00:31:52,520 Speaker 2: a sort of premise of some of the cutting. It 528 00:31:52,560 --> 00:31:56,240 Speaker 2: doesn't sound like to me when you described to be 529 00:31:56,280 --> 00:31:59,600 Speaker 2: a less in twenty nineteen through twenty twenty three, that 530 00:31:59,680 --> 00:32:03,920 Speaker 2: this is an agency that was just you know, larded up, 531 00:32:03,960 --> 00:32:06,640 Speaker 2: but had plenty of plenty of fat that can be 532 00:32:06,720 --> 00:32:07,520 Speaker 2: trimmed off. 533 00:32:07,520 --> 00:32:11,160 Speaker 5: No fat, no pad at all. But you know, this 534 00:32:11,240 --> 00:32:14,239 Speaker 5: is the age old conflict right between the entrepreneur and 535 00:32:14,280 --> 00:32:18,840 Speaker 5: the accountant. The entrepreneur always looking for innovation and changed 536 00:32:18,880 --> 00:32:22,600 Speaker 5: and higher return on investment, and the accountant is always 537 00:32:22,600 --> 00:32:25,560 Speaker 5: looking for waste and abuse. Are we using our pencils 538 00:32:25,640 --> 00:32:28,720 Speaker 5: until they are only three inches long? I think that's 539 00:32:28,720 --> 00:32:30,960 Speaker 5: a fruitful thing. I think you have to have both 540 00:32:30,960 --> 00:32:35,120 Speaker 5: of those forces working all the time and overtime. Right now. 541 00:32:35,360 --> 00:32:37,880 Speaker 5: I don't doubt for one second that a lot of 542 00:32:37,920 --> 00:32:42,400 Speaker 5: the federal government could use a thorough scrubbing on the 543 00:32:42,440 --> 00:32:46,720 Speaker 5: things that Doge is looking at. The statistical system has unfortunately, 544 00:32:46,720 --> 00:32:50,720 Speaker 5: in my opinion, but fortunately now been through that scrubbing 545 00:32:50,760 --> 00:32:55,040 Speaker 5: over the past fifteen years, no real budgets, and yet 546 00:32:55,120 --> 00:32:59,480 Speaker 5: an increase in responsibility. So efficiencies have been gained there 547 00:32:59,520 --> 00:33:02,800 Speaker 5: just from the brutality of living year after year after 548 00:33:02,880 --> 00:33:05,719 Speaker 5: year with the same dollar amount while your costs are 549 00:33:05,720 --> 00:33:09,040 Speaker 5: going up while inflation is changing. Don't mind that because 550 00:33:09,480 --> 00:33:13,520 Speaker 5: efficiencies can occur, we can do more with fewer dollars. 551 00:33:13,280 --> 00:33:17,080 Speaker 5: That's okay. Innovation, on the other hand, has to be 552 00:33:17,160 --> 00:33:19,760 Speaker 5: kind of funded by your retained earnings, and we don't 553 00:33:19,760 --> 00:33:23,840 Speaker 5: have that in The statistical Congress has that, So we're 554 00:33:23,840 --> 00:33:28,040 Speaker 5: not getting the dollars necessary to innovate and secure the future. 555 00:33:28,080 --> 00:33:30,400 Speaker 5: That's the problem I want everybody kind of focus on. 556 00:33:31,000 --> 00:33:34,280 Speaker 3: So you mentioned earlier that you had used internet data 557 00:33:34,320 --> 00:33:37,080 Speaker 3: to try to make up for the lack of a 558 00:33:37,080 --> 00:33:40,240 Speaker 3: certain data point or a certain response from the surveys, 559 00:33:40,440 --> 00:33:44,240 Speaker 3: and I'm curious, there is the sense nowadays that everything 560 00:33:44,280 --> 00:33:49,600 Speaker 3: we do is tracked and recorded somewhere. Could you potentially 561 00:33:49,800 --> 00:33:53,880 Speaker 3: use some of that type of data instead of voluntarily 562 00:33:53,960 --> 00:33:56,240 Speaker 3: reported responses. 563 00:33:57,160 --> 00:33:59,800 Speaker 5: So to an extent you can use that. If you 564 00:33:59,840 --> 00:34:04,120 Speaker 5: have an unambiguous signal, and you can capture that unambiguous 565 00:34:04,160 --> 00:34:07,680 Speaker 5: signal month after month, why not capture it? Why not capture, 566 00:34:07,720 --> 00:34:11,560 Speaker 5: for example, certain pay bands or other things that are 567 00:34:11,640 --> 00:34:15,040 Speaker 5: happening in the labor force, or with wages, or with 568 00:34:15,160 --> 00:34:20,640 Speaker 5: working conditions. The restraint there is that not everything we 569 00:34:20,719 --> 00:34:24,759 Speaker 5: want to know about the world is unambiguously signaled every month. 570 00:34:25,239 --> 00:34:28,680 Speaker 5: And I go back to this seemingly easy question, are 571 00:34:28,760 --> 00:34:32,080 Speaker 5: you working or looking for work? You would be surprised 572 00:34:32,120 --> 00:34:34,640 Speaker 5: at the number of people who say I'm working, But 573 00:34:34,680 --> 00:34:39,640 Speaker 5: then we make that query, did you work for pay? No? No, okay, 574 00:34:40,160 --> 00:34:43,479 Speaker 5: I did the dishes? Okay, fine, Well that doesn't see 575 00:34:44,120 --> 00:34:47,800 Speaker 5: in the mind of the respondent, work is defined differently 576 00:34:47,840 --> 00:34:52,080 Speaker 5: than it needs to be defined in the statistics. Are 577 00:34:52,120 --> 00:34:55,000 Speaker 5: you looking for work, Well, yeah, I'm looking for work. 578 00:34:55,040 --> 00:34:58,919 Speaker 5: When did you look for work last year? Okay? See, 579 00:34:58,920 --> 00:35:02,759 Speaker 5: that doesn't count as the key determinant of whether or 580 00:35:02,800 --> 00:35:04,760 Speaker 5: not you're in the labor force. You're in the labor 581 00:35:04,760 --> 00:35:06,960 Speaker 5: force if you're working or looking for work in the 582 00:35:07,000 --> 00:35:12,400 Speaker 5: past four weeks. So that seemingly simple question is not 583 00:35:12,480 --> 00:35:17,520 Speaker 5: going to be unambiguously signaled by somebody answering an Internet question. 584 00:35:17,920 --> 00:35:20,480 Speaker 5: Because of the many different ways we think about our 585 00:35:20,520 --> 00:35:24,040 Speaker 5: lives and about what work is. We find this particularly 586 00:35:24,120 --> 00:35:26,879 Speaker 5: true as our country becomes I think it's a good thing. 587 00:35:27,320 --> 00:35:30,600 Speaker 5: More and more culturally diverse. People come in with very 588 00:35:30,600 --> 00:35:33,960 Speaker 5: different views of what is work of, what is pay of, 589 00:35:34,200 --> 00:35:38,920 Speaker 5: what is a family of, what is a household? And 590 00:35:39,200 --> 00:35:42,720 Speaker 5: we have to work harder and harder and harder to 591 00:35:42,760 --> 00:35:47,480 Speaker 5: make sure that their responses fit this continuous since nineteen 592 00:35:47,560 --> 00:35:51,080 Speaker 5: forty three, continuous stream of data that allows us to 593 00:35:51,080 --> 00:35:55,040 Speaker 5: do these wonderful time series analyzes. That's the issue tracy 594 00:35:55,080 --> 00:35:59,080 Speaker 5: with using Internet data. Some of it can be used 595 00:35:59,400 --> 00:36:02,280 Speaker 5: and blended in, some of it from the private sector 596 00:36:02,280 --> 00:36:05,600 Speaker 5: can be used and blended in, but some cannot. You 597 00:36:06,040 --> 00:36:09,520 Speaker 5: still have to have that survey instrument out there, asking 598 00:36:09,560 --> 00:36:11,600 Speaker 5: the hard questions, doing the follow ups. 599 00:36:12,080 --> 00:36:14,120 Speaker 2: You know, there are some people who simply do not 600 00:36:14,320 --> 00:36:18,399 Speaker 2: believe these statistical agencies are honest, or they believe they 601 00:36:18,520 --> 00:36:22,719 Speaker 2: you know, they do not trust when academic economists explain 602 00:36:22,920 --> 00:36:26,359 Speaker 2: why data gets revised a year later or something like that, 603 00:36:26,440 --> 00:36:31,080 Speaker 2: and they assume that there is political influence of some sort, 604 00:36:31,160 --> 00:36:32,600 Speaker 2: or they believe it and you know, it goes back 605 00:36:32,680 --> 00:36:36,560 Speaker 2: years and I remember two thousand and twelve years ago, 606 00:36:36,719 --> 00:36:39,600 Speaker 2: Jack Welch, the Chicago guys will do anything, can't believe 607 00:36:39,600 --> 00:36:42,920 Speaker 2: these jobs numbers, et cetera. You were appointed in twenty 608 00:36:43,000 --> 00:36:46,080 Speaker 2: nineteen by President Trump and you crossed over to Biden. 609 00:36:46,200 --> 00:36:48,920 Speaker 2: What do you say to people who do not believe 610 00:36:49,560 --> 00:36:50,800 Speaker 2: that they can trust these numbers? 611 00:36:52,000 --> 00:36:57,200 Speaker 5: Well, I can't dislaunch their suspicions without going to some 612 00:36:57,360 --> 00:37:02,640 Speaker 5: factual basis. So I take them through the simplest revision process, 613 00:37:02,640 --> 00:37:04,520 Speaker 5: which is the jobs revision that you know. We had 614 00:37:04,520 --> 00:37:07,920 Speaker 5: an eight hundred and eighteen thousand preliminary estimate of a 615 00:37:08,000 --> 00:37:11,240 Speaker 5: decrease in employment that was announced last August, and everybody 616 00:37:11,400 --> 00:37:14,279 Speaker 5: President Trump was running for at the time, and he 617 00:37:14,360 --> 00:37:17,680 Speaker 5: just said, look at this, it is totally dishonest. Bls okay. 618 00:37:17,760 --> 00:37:23,200 Speaker 5: So this happens every year. We compare our numbers to 619 00:37:24,160 --> 00:37:27,120 Speaker 5: a sampling frame with all the people who work in 620 00:37:27,160 --> 00:37:31,000 Speaker 5: the entire country, and we say, is our estimate of 621 00:37:31,040 --> 00:37:33,719 Speaker 5: total employment based on a much smaller sample than all 622 00:37:33,760 --> 00:37:36,719 Speaker 5: the firms. Is it accurate? And when we go up 623 00:37:36,760 --> 00:37:40,000 Speaker 5: there and we compare it every March to this total universe, 624 00:37:40,440 --> 00:37:43,359 Speaker 5: we sometimes find we're spot on, you know, less than 625 00:37:43,400 --> 00:37:46,359 Speaker 5: one tenth of a percent off, and sometimes we find 626 00:37:46,400 --> 00:37:49,080 Speaker 5: we're as much as two percent or three percent off. 627 00:37:49,600 --> 00:37:53,080 Speaker 5: That we're off more often when the economy is either 628 00:37:53,280 --> 00:37:57,360 Speaker 5: diving into recession or growing rapidly out of a recession, 629 00:37:57,960 --> 00:38:00,799 Speaker 5: or has a period of really bad time happening. So 630 00:38:01,080 --> 00:38:03,520 Speaker 5: when I take them through the revision process, they usually 631 00:38:03,560 --> 00:38:05,480 Speaker 5: come out and say, you know, I never knew that, 632 00:38:05,719 --> 00:38:09,160 Speaker 5: And so the next time they think about, oh, the 633 00:38:09,160 --> 00:38:12,200 Speaker 5: BLS is lying, they'll have that in their mind that really, 634 00:38:12,960 --> 00:38:15,240 Speaker 5: we did this every year it is the same way, 635 00:38:15,800 --> 00:38:17,320 Speaker 5: and we're pretty good with those numbers. 636 00:38:17,960 --> 00:38:21,440 Speaker 3: So I want to ask you about qualitative adjustments because 637 00:38:21,440 --> 00:38:24,320 Speaker 3: I find these so interesting. And the example I used 638 00:38:24,680 --> 00:38:28,239 Speaker 3: was fridges that now have I don't know, new and 639 00:38:28,360 --> 00:38:32,000 Speaker 3: interesting features. And my understanding is that if a company 640 00:38:32,040 --> 00:38:36,080 Speaker 3: is selling a basic refrigerator for like one thousand dollars, 641 00:38:36,400 --> 00:38:39,400 Speaker 3: and then the next year it sells this new advanced 642 00:38:39,480 --> 00:38:43,480 Speaker 3: refrigerator for one thousand, one hundred and fifty, and then 643 00:38:43,520 --> 00:38:46,400 Speaker 3: it's also spending additional money, so like an extra hundred 644 00:38:46,480 --> 00:38:50,520 Speaker 3: or whatever to make the more advanced fridge. In cases 645 00:38:50,560 --> 00:38:54,920 Speaker 3: like that, the BLS would use a qualitative adjustment and 646 00:38:54,960 --> 00:38:58,160 Speaker 3: then the year on year PPI would be something like, 647 00:38:58,760 --> 00:39:01,120 Speaker 3: I don't know, it would be five percent instead of 648 00:39:01,560 --> 00:39:04,680 Speaker 3: the fifteen percent change in the actual price. How do 649 00:39:04,719 --> 00:39:08,040 Speaker 3: you actually go about making those qualitative adjustments? And I 650 00:39:08,040 --> 00:39:11,200 Speaker 3: imagine they must have been getting harder as things become 651 00:39:11,640 --> 00:39:13,240 Speaker 3: more technologically complex. 652 00:39:14,719 --> 00:39:17,279 Speaker 5: Well, you're absolutely right, it is very difficult to do that, 653 00:39:17,400 --> 00:39:20,600 Speaker 5: but we do a lot of training. A lot of 654 00:39:20,600 --> 00:39:24,480 Speaker 5: people don't know that. At the National Headquarters there is 655 00:39:24,520 --> 00:39:27,800 Speaker 5: a training suite. I think it's on the second second floor, 656 00:39:27,800 --> 00:39:30,080 Speaker 5: at least be before we moved out of that building. 657 00:39:30,719 --> 00:39:34,040 Speaker 5: And so people from the regions who are the CPI 658 00:39:34,239 --> 00:39:38,320 Speaker 5: and PPI field teams that go out and do the surveying, 659 00:39:38,560 --> 00:39:41,279 Speaker 5: they come to the bl AS for training. And in 660 00:39:41,320 --> 00:39:46,640 Speaker 5: these training rooms our kitchens, there are grocery store aisles. 661 00:39:46,719 --> 00:39:50,200 Speaker 5: We've recreated kind of the inventory you might find in 662 00:39:50,239 --> 00:39:53,319 Speaker 5: a grocery store or a warehouse, and we will train 663 00:39:53,440 --> 00:39:56,600 Speaker 5: people from time to time on the changes in the 664 00:39:56,680 --> 00:40:00,200 Speaker 5: items in the CPI, the two hundred plus items the 665 00:40:00,200 --> 00:40:04,400 Speaker 5: CPI or the items that we're surveying at the producer 666 00:40:04,440 --> 00:40:09,000 Speaker 5: price level. So if there is a change all the 667 00:40:09,000 --> 00:40:11,560 Speaker 5: way from the number of potato chips in a bag, 668 00:40:11,960 --> 00:40:14,279 Speaker 5: which is a very important thing. And when we do 669 00:40:14,320 --> 00:40:17,280 Speaker 5: the we will look at fast foods and potato chips 670 00:40:17,280 --> 00:40:20,239 Speaker 5: and so forth, all the way to the technology involved 671 00:40:20,400 --> 00:40:24,719 Speaker 5: in diamond cutting. We will be training people to observe 672 00:40:25,120 --> 00:40:29,560 Speaker 5: the change and work that into their evaluation of the 673 00:40:29,600 --> 00:40:32,200 Speaker 5: product when they're in the store, when they're in the warehouse. 674 00:40:32,600 --> 00:40:37,040 Speaker 5: So the field person will literally pick up a jar 675 00:40:37,280 --> 00:40:40,520 Speaker 5: of if I could say pringles, right, and they'll say, well, 676 00:40:40,600 --> 00:40:43,680 Speaker 5: last month I had thirty six springles in here, and 677 00:40:44,040 --> 00:40:46,720 Speaker 5: it's this month it's the same price, but we only 678 00:40:46,760 --> 00:40:49,040 Speaker 5: have thirty two pringles here. 679 00:40:49,560 --> 00:40:52,000 Speaker 3: Wow. True regularity. 680 00:40:52,760 --> 00:40:55,400 Speaker 5: Yeah, so that that fits into. 681 00:40:55,360 --> 00:40:58,960 Speaker 4: Now price perles keep going exactly. 682 00:40:59,640 --> 00:41:02,800 Speaker 5: But it has to be that way because what happened 683 00:41:02,840 --> 00:41:05,239 Speaker 5: is the retailer will keep the or the producer will 684 00:41:05,280 --> 00:41:07,560 Speaker 5: keep the price at the same level, but decrease the 685 00:41:07,600 --> 00:41:10,799 Speaker 5: item count in the in the product bag, and that 686 00:41:10,880 --> 00:41:14,040 Speaker 5: means that the product has actually gone up in price, 687 00:41:14,320 --> 00:41:18,480 Speaker 5: not have the same price. Let me just say one 688 00:41:18,520 --> 00:41:21,799 Speaker 5: more thing, great question for Tracy. We have. That's the 689 00:41:21,880 --> 00:41:24,920 Speaker 5: reason why we can't go to the internet and get 690 00:41:24,960 --> 00:41:28,160 Speaker 5: all of our prices because when you go to the internet, 691 00:41:28,280 --> 00:41:31,279 Speaker 5: you can't sometimes see the quality adjustments that are there. 692 00:41:31,360 --> 00:41:34,320 Speaker 5: You can't see that the fact that the hot dogs 693 00:41:34,440 --> 00:41:37,680 Speaker 5: are just a little shorter than they used to be, 694 00:41:38,200 --> 00:41:40,719 Speaker 5: or the apples a little smaller than they used to be. 695 00:41:41,080 --> 00:41:43,280 Speaker 5: I don't know about the apple thing, but the shorter 696 00:41:43,560 --> 00:41:46,239 Speaker 5: hot dogs is definite the case. So you have to 697 00:41:46,719 --> 00:41:49,319 Speaker 5: we train these people. They're highly trained. That's they're also 698 00:41:49,440 --> 00:41:53,120 Speaker 5: very expensive because they're highly trained people and they can 699 00:41:53,160 --> 00:41:56,160 Speaker 5: see and know when to look and when to check 700 00:41:56,400 --> 00:42:00,799 Speaker 5: for quality improvements. Electronics definitely, you know, but a lot 701 00:42:00,840 --> 00:42:04,600 Speaker 5: of times the producers of the electronics will heavily advertise 702 00:42:04,680 --> 00:42:07,360 Speaker 5: the changes, make it known to everybody, because that's what 703 00:42:07,400 --> 00:42:10,560 Speaker 5: you're selling new and improved. It's all of these other 704 00:42:10,640 --> 00:42:14,240 Speaker 5: items that it's more subtle. And you know, housing, Housing 705 00:42:14,320 --> 00:42:17,640 Speaker 5: is a big deal because the house is more valuable 706 00:42:17,760 --> 00:42:20,399 Speaker 5: if it has improvements in it, and some of those 707 00:42:20,440 --> 00:42:25,239 Speaker 5: improvements are completely invisible. So we're we're very conscious of 708 00:42:25,280 --> 00:42:29,160 Speaker 5: the fact that quality governs the price structure. 709 00:42:30,320 --> 00:42:33,520 Speaker 2: Bill Beach, that was a fascinating conversation. I feel like 710 00:42:33,560 --> 00:42:36,399 Speaker 2: some of these areas like even just like I'm talking 711 00:42:36,400 --> 00:42:39,319 Speaker 2: about the art of actually conducting an interview you see 712 00:42:39,320 --> 00:42:42,600 Speaker 2: if someone's in the labor force, really fascinating stuff. 713 00:42:42,880 --> 00:42:44,879 Speaker 4: Really appreciate you coming on. There's something we want. 714 00:42:45,040 --> 00:42:47,399 Speaker 2: We wanted to talk to you, to talk to someone 715 00:42:47,480 --> 00:42:49,880 Speaker 2: long time as so appreciate you joining outlines. 716 00:42:50,840 --> 00:42:52,839 Speaker 5: It's been it's been a pleasure. Thank you very much. 717 00:42:53,080 --> 00:42:53,879 Speaker 3: Thank you so much. 718 00:42:53,920 --> 00:43:09,719 Speaker 4: That was great, Tracy. I thought that was great. 719 00:43:10,160 --> 00:43:12,480 Speaker 3: I'm so glad we finally did that one. Yeah, a 720 00:43:12,480 --> 00:43:14,480 Speaker 3: couple points. So, first of all, I did not know 721 00:43:14,520 --> 00:43:16,160 Speaker 3: that hot dogs have been getting shorter. 722 00:43:16,280 --> 00:43:16,759 Speaker 4: Me neither. 723 00:43:17,239 --> 00:43:20,080 Speaker 3: And it's being incorporated into the inflation data, So all 724 00:43:21,160 --> 00:43:24,239 Speaker 3: the people complaining about shrink flation, I guess you know 725 00:43:24,360 --> 00:43:28,160 Speaker 3: it is mitigated. Yeah, exactly. And then the other thing 726 00:43:28,200 --> 00:43:30,160 Speaker 3: I would say is I thought the point Bill was 727 00:43:30,200 --> 00:43:34,319 Speaker 3: making about the loss of granularity in the data was 728 00:43:34,360 --> 00:43:37,400 Speaker 3: really important. So the idea of getting to like the 729 00:43:37,560 --> 00:43:41,320 Speaker 3: tails of the distribution or certain minorities. And the reason 730 00:43:41,360 --> 00:43:43,919 Speaker 3: I say that is because we've been seeing a lot 731 00:43:43,960 --> 00:43:48,600 Speaker 3: of regional and social variation in a lot of these 732 00:43:48,680 --> 00:43:53,400 Speaker 3: consumer surveys. Right, So obviously people who are poorer have 733 00:43:53,520 --> 00:43:57,560 Speaker 3: been feeling terrible during the days of high inflation, and 734 00:43:57,600 --> 00:44:00,160 Speaker 3: people who are rich feel, you know, pretty good, good 735 00:44:00,239 --> 00:44:04,440 Speaker 3: but also inflation in Florida has been higher than elsewhere 736 00:44:04,520 --> 00:44:07,319 Speaker 3: in the country. So I think it does become more 737 00:44:07,360 --> 00:44:10,359 Speaker 3: important to get really really specific with some of these 738 00:44:10,400 --> 00:44:13,200 Speaker 3: statistics as we see those differences increase. 739 00:44:13,560 --> 00:44:18,279 Speaker 2: It's really interesting to me to think about government high 740 00:44:18,320 --> 00:44:22,239 Speaker 2: quality government data as like this public good and try 741 00:44:22,280 --> 00:44:25,920 Speaker 2: to imagine infrastructure, right infrastructure, and try to imagine the 742 00:44:26,000 --> 00:44:31,440 Speaker 2: amount of economic activity that exists because this thing is 743 00:44:31,560 --> 00:44:33,960 Speaker 2: offered for free that we don't that people don't have 744 00:44:34,000 --> 00:44:37,160 Speaker 2: to pay for. And obviously in the investing world there's 745 00:44:37,160 --> 00:44:39,120 Speaker 2: a tremendous amount of interest in all this, but it's 746 00:44:39,160 --> 00:44:41,719 Speaker 2: obviously not just you know, the investing world, and so 747 00:44:41,840 --> 00:44:44,440 Speaker 2: all of these questions, whether you're starting a business or 748 00:44:44,480 --> 00:44:47,799 Speaker 2: whatever it is, you know, on some level you can 749 00:44:47,880 --> 00:44:52,560 Speaker 2: do because there is consistent, trustworthy data. And the thought 750 00:44:52,600 --> 00:44:56,359 Speaker 2: of like that going away, and what I imagine what happened is, yeah, 751 00:44:56,480 --> 00:44:59,840 Speaker 2: sure you'd have like private versions of varying quality that 752 00:45:00,200 --> 00:45:02,640 Speaker 2: try to replace it, and that exists today. You know, 753 00:45:02,680 --> 00:45:05,680 Speaker 2: there's private measures of inflation, et cetera. But like that 754 00:45:05,719 --> 00:45:08,399 Speaker 2: would like start to deteriorate the idea of like there 755 00:45:08,440 --> 00:45:11,279 Speaker 2: being a gold standard, and so then you hear it's like, oh, 756 00:45:11,360 --> 00:45:13,200 Speaker 2: here's like a oh, we needed to find a few 757 00:45:13,200 --> 00:45:17,440 Speaker 2: more million dollars to pay the budgets of the survey collectors. 758 00:45:17,680 --> 00:45:20,880 Speaker 2: Like how many billions are riding on that few extra millions? 759 00:45:20,800 --> 00:45:23,800 Speaker 2: Oh yeah, right, like how many hundreds of billions in activity? 760 00:45:23,960 --> 00:45:28,160 Speaker 2: So it's just really interesting. And then his thing explanation 761 00:45:28,239 --> 00:45:31,200 Speaker 2: at the end, why like survey collection for something like 762 00:45:31,239 --> 00:45:34,839 Speaker 2: this is it makes sense to do as a trained job, right, 763 00:45:34,920 --> 00:45:37,480 Speaker 2: and even like the subjectivity of like are you working 764 00:45:37,480 --> 00:45:39,680 Speaker 2: and what does that mean and et cetera. It's really 765 00:45:39,719 --> 00:45:43,560 Speaker 2: interesting to think about why it's not trivial to just 766 00:45:43,600 --> 00:45:45,440 Speaker 2: send out a survey or send out a high school 767 00:45:45,560 --> 00:45:47,279 Speaker 2: or send out a volunteer or something like that. 768 00:45:47,360 --> 00:45:50,160 Speaker 3: Absolutely. Also, I just love the idea of someone at 769 00:45:50,160 --> 00:45:53,040 Speaker 3: the BLS counting how many potato chips are in a bag? 770 00:45:53,200 --> 00:45:57,000 Speaker 2: I know, you know what I'm imagining, like I don't know, 771 00:45:57,040 --> 00:45:59,560 Speaker 2: like someone with like a monocle, yeah, or something like. 772 00:45:59,600 --> 00:46:04,319 Speaker 3: The magnifying class examining shifts seeing how big they are. 773 00:46:04,400 --> 00:46:07,000 Speaker 2: We have we're having the same image in our minds, 774 00:46:07,000 --> 00:46:07,839 Speaker 2: that's sure right now? 775 00:46:08,000 --> 00:46:09,400 Speaker 3: Yeah, all right, shall we leave it there? 776 00:46:09,480 --> 00:46:10,359 Speaker 4: Yeah, let's leave it there. 777 00:46:10,440 --> 00:46:13,279 Speaker 3: This has been another episode of the Authoughts podcast. I'm 778 00:46:13,280 --> 00:46:15,879 Speaker 3: Tracy Alloway. You can follow me at Tracy. 779 00:46:15,640 --> 00:46:18,399 Speaker 2: Alloway and I'm Jill Wisenthal. You can follow me at 780 00:46:18,400 --> 00:46:22,279 Speaker 2: the Stalwart. Follow our producers Kerman Rodriguez at Kerman armand 781 00:46:22,360 --> 00:46:25,680 Speaker 2: Dashill Bennett at Dashbot and kil Brooks at Kale Brooks. 782 00:46:25,800 --> 00:46:28,160 Speaker 2: For more odd Laws content, go to Bloomberg dot com 783 00:46:28,160 --> 00:46:31,000 Speaker 2: slash odd Lots, where we have all of our episodes 784 00:46:31,040 --> 00:46:33,239 Speaker 2: in the daily newsletter and you can chat about all 785 00:46:33,280 --> 00:46:35,960 Speaker 2: of these topics twenty four to seven in our discord 786 00:46:36,000 --> 00:46:38,040 Speaker 2: Discord dot gg slash od. 787 00:46:37,840 --> 00:46:40,440 Speaker 3: Lots And if you enjoy odd Lots, if you like 788 00:46:40,520 --> 00:46:43,520 Speaker 3: it when we talk about how these statistics sausage actually 789 00:46:43,520 --> 00:46:46,400 Speaker 3: gets made, then please leave us a positive review on 790 00:46:46,440 --> 00:46:49,839 Speaker 3: your favorite podcast platform. And remember, if you are a 791 00:46:49,880 --> 00:46:53,239 Speaker 3: Bloomberg subscriber, you can listen to all of our episodes 792 00:46:53,360 --> 00:46:55,839 Speaker 3: absolutely ad free. All you need to do is find 793 00:46:55,880 --> 00:46:59,640 Speaker 3: the Bloomberg channel on Apple Podcasts and follow the instructions there. 794 00:47:00,160 --> 00:47:00,960 Speaker 3: Thanks for listening.