1 00:00:00,500 --> 00:00:05,443 The Bureau of Labor Statistics rec released a July Consumer Price Index 2 00:00:06,024 --> 00:00:09,336 on August twelfth, which is the day that we're recording this at 8 3 00:00:09,336 --> 00:00:16,050 30 in the morning. The headline number came in at 3.4% year over year inflation. 4 00:00:16,050 --> 00:00:20,524 Core CPI, excluding food and energy, 5 00:00:20,524 --> 00:00:23,596 hit two point five percent. Markets moved. 6 00:00:23,596 --> 00:00:25,717 The Federal Reserve was watching. 7 00:00:25,717 --> 00:00:28,700 Or at least that's what we're supposed to believe. 8 00:00:28,700 --> 00:00:29,700 But here's what 9 00:00:29,512 --> 00:00:31,532 made the number actually interesting. 10 00:00:31,532 --> 00:00:37,374 Nearly two thirds of that monthly increase came from one category shelter, 11 00:00:37,374 --> 00:00:41,055 not gas, not groceries, shelter. 12 00:00:41,055 --> 00:00:45,565 A single component of the index accounted for the bulk of the signal 13 00:00:45,565 --> 00:00:49,277 the Fed is using to decide whether you can afford to borrow money, 14 00:00:49,277 --> 00:00:52,117 whether your paycheck needs to stretch further, 15 00:00:52,117 --> 00:00:58,629 whether the economy is running too hot or too cold or on the process of cooling. 16 00:00:58,993 --> 00:01:04,667 Except shelter in the CPI isn't measured the way you pay for shelter. 17 00:01:04,667 --> 00:01:08,009 It's not your mortgage. It's not your actual rent cost. 18 00:01:08,009 --> 00:01:11,691 It's something called owner's equivalent rent. 19 00:01:11,691 --> 00:01:16,815 A hypothetical number that represents the rent homeowners think they could charge 20 00:01:16,815 --> 00:01:21,558 themselves if they rented their own house to themselves. 21 00:01:21,558 --> 00:01:25,000 Number one actually pays it. 22 00:01:25,000 --> 00:01:26,841 The Fed watches this one number. 23 00:01:26,841 --> 00:01:28,162 Your rent is 24 00:01:28,381 --> 00:01:31,708 And your mortgage bill lives somewhere else entirely. 25 00:01:50,573 --> 00:01:52,715 Hey there, I am Joshua. 26 00:01:53,215 --> 00:01:54,215 And I'm Will. 27 00:01:54,529 --> 00:01:56,496 And this is the overlap. Will's back, 28 00:01:56,496 --> 00:01:57,496 you guys. 29 00:01:57,686 --> 00:01:58,831 We're glad to be back, Joshua. 30 00:01:58,831 --> 00:01:59,831 Thank you. 31 00:01:59,980 --> 00:02:04,414 Glad to be with ya. Why don't you tell the our our guests, 32 00:02:04,414 --> 00:02:06,467 our wonderful audience, our beloved, 33 00:02:06,467 --> 00:02:10,591 esteemed, good looking, talented audience what we're talking about today. 34 00:02:11,091 --> 00:02:12,562 Well as you probably heard in the intro, 35 00:02:12,562 --> 00:02:14,414 we're talking today about the CPI, 36 00:02:14,414 --> 00:02:17,247 or consumer price index report that landed this morning, 37 00:02:17,247 --> 00:02:21,781 and why the inflation number everyone heard isn't actually the number that's driving 38 00:02:21,781 --> 00:02:22,781 policy. 39 00:02:23,090 --> 00:02:26,801 Right. So yesterday, like I said at the at the the intro, 40 00:02:26,801 --> 00:02:32,322 the Bureau of Labor Statistics puts out July's inflation data three point four 41 00:02:32,322 --> 00:02:37,524 percent year over year. That's the headline that everyone saw. 42 00:02:37,524 --> 00:02:43,125 But if you actually look at what moved that number and like actually look at it, 43 00:02:43,125 --> 00:02:45,486 it's not just the press release. 44 00:02:45,486 --> 00:02:50,227 You find something pretty crazy about how the government measures inflation. 45 00:02:50,749 --> 00:02:54,097 And how far that measurement has actually drifted from what people 46 00:02:54,097 --> 00:02:55,130 are really feeling? 47 00:02:55,630 --> 00:02:58,887 And why does this matter beyond just being a data nerd complaint? 48 00:02:59,387 --> 00:03:02,018 Because the Federal Reserve uses that number, 49 00:03:02,018 --> 00:03:05,070 or at least some version of this number that we're not allowed to know, 50 00:03:05,070 --> 00:03:09,232 like a credit score, to decide what interest rates look like. 51 00:03:09,232 --> 00:03:11,923 And that decision ripples through everything. 52 00:03:11,923 --> 00:03:13,824 Your mortgage rate, your credit cards, 53 00:03:13,824 --> 00:03:18,187 your car loans, whether employees are our employers, 54 00:03:18,187 --> 00:03:20,818 sorry, are hiring, whether they're they're cutting jobs, 55 00:03:20,818 --> 00:03:25,821 they're laying people off. It all traces back to how they're measuring inflation. 56 00:03:25,821 --> 00:03:28,472 And the way they're measuring it is to be 57 00:03:29,114 --> 00:03:32,844 charitable I guess, not connected to how any of us actually live. 58 00:03:33,803 --> 00:03:37,942 Surprise, surprise. But you mean this is like the number that will help determine 59 00:03:37,942 --> 00:03:41,008 when you go to the grocery store and everything's ten percent more expensive than 60 00:03:41,008 --> 00:03:43,213 last month, but the government says inflation has eased? 61 00:03:43,937 --> 00:03:47,610 Exactly that. And we talked about it in it last episode actually. 62 00:03:47,610 --> 00:03:49,091 If you if you didn't listen to it, 63 00:03:49,091 --> 00:03:53,094 go back and listen to the episode ~ talking about democratic socialism. 64 00:03:53,094 --> 00:03:59,590 But it this is kind of a a specific reason that shelter is measured the 65 00:03:59,590 --> 00:04:02,354 way that it is. And it really goes back about forty years. 66 00:04:02,354 --> 00:04:04,924 It always goes back forty years because that's what we do, 67 00:04:04,924 --> 00:04:08,357 right? We connect things that are happening now to things that happened in the past. 68 00:04:08,357 --> 00:04:13,611 And there's a specific reason that the Federal Reserve is using a different 69 00:04:13,805 --> 00:04:16,699 inflation metric than the one everyone actually talks about. 70 00:04:16,699 --> 00:04:18,913 And that's kind of deliberate too. 71 00:04:18,913 --> 00:04:21,477 And and both of these things are sitting on top of each other right now 72 00:04:21,477 --> 00:04:27,165 in a way that makes the official story and the kind of lived experience really 73 00:04:27,286 --> 00:04:28,948 incomprehensible to each other. 74 00:04:29,448 --> 00:04:31,067 Well it's a good thing we're here to pull that apart. 75 00:04:31,953 --> 00:04:38,117 Right, exactly. So before we were even born in 1983, 76 00:04:38,117 --> 00:04:40,779 before I was born. Well, you were born before then, 77 00:04:39,667 --> 00:04:42,364 Hm. Ancient times. Ancient times. 78 00:04:40,779 --> 00:04:46,013 right? 82. Back in the day. It was a Tuesday. 79 00:04:42,364 --> 00:04:44,892 Yes. It was a harsh life. 80 00:04:47,274 --> 00:04:51,207 But no, in in 1983, the Bureau of Labor Statistics made a decision about 81 00:04:51,207 --> 00:04:56,580 how to measure housing costs inside the consumer price index. 82 00:04:56,580 --> 00:04:59,763 Now, if you don't know what the consumer price index is. 83 00:05:00,447 --> 00:05:03,328 at the time the way they did it was kind of straightforward. 84 00:05:03,328 --> 00:05:07,950 They just measured the actual home sale prices and you wanted 85 00:05:07,950 --> 00:05:10,181 to know what homeownership cost. 86 00:05:10,181 --> 00:05:13,632 You you tracked that, like what people paid for houses, 87 00:05:13,632 --> 00:05:18,174 which makes sense, right? It sounds like the most obvious way to measure 88 00:05:18,895 --> 00:05:19,895 the cost of housing. 89 00:05:20,345 --> 00:05:21,955 So of course they stopped doing it that way. 90 00:05:22,455 --> 00:05:24,536 Right, and for very different reasons. 91 00:05:24,536 --> 00:05:27,357 So they did stop doing that. They they switched to something called 92 00:05:27,357 --> 00:05:31,669 the owner's equivalent rent. The idea is that instead of measuring what 93 00:05:31,669 --> 00:05:36,971 a house actually costs, you measure what that house would cost to rent 94 00:05:37,471 --> 00:05:40,192 if the owner rented it out for the price 95 00:05:39,762 --> 00:05:40,762 Yeah. 96 00:05:40,192 --> 00:05:41,559 that they felt was reasonable. 97 00:05:41,559 --> 00:05:43,354 I I don't even know how they measure this, 98 00:05:43,354 --> 00:05:49,396 but then you're basically comparing the cost of owning to the cost of renting. 99 00:05:50,086 --> 00:05:51,588 And you're measuring the difference. 100 00:05:51,588 --> 00:05:57,705 So in theory, if owning a house becomes more expensive relative to renting, 101 00:05:57,705 --> 00:06:02,281 owners would start selling and landlording those houses out, 102 00:06:02,281 --> 00:06:05,235 which would theoretically drive equilibrium. 103 00:06:05,235 --> 00:06:08,609 It's not exactly an elegant theory, 104 00:06:08,609 --> 00:06:09,609 but it is a theory. 105 00:06:09,900 --> 00:06:11,982 Right, but nobody actually pays the owner's equivalent rent, 106 00:06:11,982 --> 00:06:14,143 right? I mean we're talking about a hypothetical number. 107 00:06:14,143 --> 00:06:18,045 It's it's it's this seems pretty far removed from reality. 108 00:06:18,481 --> 00:06:19,481 Exactly, 109 00:06:19,011 --> 00:06:22,102 because i it's supposed to be some sort of balance. 110 00:06:22,102 --> 00:06:25,014 But number one, i it's hypothetical, 111 00:06:25,014 --> 00:06:29,636 right? It's a number constructed from surveys where they ask people 112 00:06:29,636 --> 00:06:33,017 who actually own homes what they think their house would rent for. 113 00:06:33,017 --> 00:06:38,549 Gabriel Chottero Reich at Harvard has done a lot of research on this. 114 00:06:38,549 --> 00:06:41,941 He he found that the the BLS, that's the Bureau of Labor Statistics 115 00:06:41,941 --> 00:06:44,222 for those TLA people out there, 116 00:06:44,222 --> 00:06:47,433 three letter acronyms, it's constructing 117 00:06:47,696 --> 00:06:52,569 A phantom rent that doesn't actually exist anywhere in our actual economy. 118 00:06:52,569 --> 00:06:55,980 So your mortgage payment doesn't actually get measured, 119 00:06:55,980 --> 00:06:59,502 even though it should be. Your property taxes don't actually get measured, 120 00:06:59,502 --> 00:07:01,393 and let's be honest, it really should be. 121 00:07:01,393 --> 00:07:05,445 The actual price you paid for your house doesn't even get measured. 122 00:07:05,445 --> 00:07:10,638 What gets measured is a survey that respondents guess about 123 00:07:10,638 --> 00:07:15,601 a hypothetical rental situation that will never actually happen because they live 124 00:07:15,601 --> 00:07:16,601 in the house. 125 00:07:16,581 --> 00:07:19,407 Not to mention that people are often deluded about what their property 126 00:07:19,407 --> 00:07:21,571 is worth or what their things would be worth to someone else. 127 00:07:21,571 --> 00:07:22,713 ~ 128 00:07:21,936 --> 00:07:22,936 exactly. 129 00:07:22,733 --> 00:07:25,498 in all sorts of directions. But why why would they do this? 130 00:07:26,265 --> 00:07:30,287 Yeah. The official reason that they give is that it gives you 131 00:07:30,287 --> 00:07:33,819 a measure that's comparable between renters and owners. 132 00:07:33,819 --> 00:07:37,161 Does that sound confusing? Because it feels confusing. 133 00:07:37,149 --> 00:07:38,149 It's yeah. 134 00:07:37,161 --> 00:07:38,301 So ~ 135 00:07:38,301 --> 00:07:41,703 a renter pays rent. An owner, 136 00:07:41,703 --> 00:07:45,726 in this this wonderful, beautiful fairy tale logic, 137 00:07:45,726 --> 00:07:50,008 should be treated as someone paying themselves rent. 138 00:07:50,008 --> 00:07:51,068 It kind of 139 00:07:51,555 --> 00:07:55,419 smooths out the volatility of actual home sale prices, 140 00:07:55,419 --> 00:07:58,331 which do in fact swing very widely. 141 00:07:58,331 --> 00:08:01,525 If you've ever bought or purchased a home or sold 142 00:08:00,917 --> 00:08:01,917 Yeah. 143 00:08:01,525 --> 00:08:04,077 a home or were trying to purchase a home, 144 00:08:04,077 --> 00:08:07,130 you know that the numbers just don't seem connected to reality. 145 00:08:07,630 --> 00:08:08,630 Right. 146 00:08:08,694 --> 00:08:10,716 It gives you something more kind of stable, 147 00:08:10,716 --> 00:08:12,787 I guess, to put inside of an index, 148 00:08:12,787 --> 00:08:15,179 which that's the goal of an index. 149 00:08:15,179 --> 00:08:17,061 It's like an average. And honestly, 150 00:08:17,061 --> 00:08:19,663 for a certain kind of economist, 151 00:08:19,663 --> 00:08:22,665 and they're sure there's at least one of you listening out there, 152 00:08:22,665 --> 00:08:25,447 it feels more theoretically pure. 153 00:08:25,447 --> 00:08:27,859 You're not really measuring asset prices, 154 00:08:27,859 --> 00:08:30,451 you're measuring the service of housing, 155 00:08:30,451 --> 00:08:32,553 like the actual roof over your head. 156 00:08:33,053 --> 00:08:36,756 But first of all, let me just say when you s when you use the word smooth 157 00:08:36,756 --> 00:08:41,000 out like volatility, ~ thanks in no small part to this podcast, 158 00:08:41,000 --> 00:08:42,742 ~ alarm bells go off in my head, 159 00:08:42,742 --> 00:08:44,293 like why would they be smoothing that out? 160 00:08:44,293 --> 00:08:46,285 Like what who else stands to gain from that? 161 00:08:46,285 --> 00:08:48,567 But I have a different a little bit different question. 162 00:08:48,567 --> 00:08:51,299 If home prices spike and the Fed wants to know whether inflation 163 00:08:51,299 --> 00:08:54,122 is happening in housing, they're not gonna see it under this new model, 164 00:08:54,122 --> 00:08:55,122 right? 165 00:08:55,116 --> 00:08:56,116 Right, they're 166 00:08:55,606 --> 00:09:00,857 not and and I do wanna be ~ I wanna be ~ not just honest but like give give true 167 00:09:00,857 --> 00:09:03,508 accounts for for someone who who crunches data, 168 00:09:03,508 --> 00:09:07,149 right, and is not necessarily worried about about the actual outcomes 169 00:09:07,149 --> 00:09:10,050 of lives and the the feelings and the people who live in the 170 00:09:10,050 --> 00:09:11,710 the numbers that they're talking about. 171 00:09:11,710 --> 00:09:16,981 It makes it easier to compute because otherwise it just it's just th you 172 00:09:16,981 --> 00:09:22,709 can draw a line through all of the data and you try to find the mean in all of it. 173 00:09:22,709 --> 00:09:24,003 It it just makes it easier. 174 00:09:24,663 --> 00:09:26,455 To give real world results. Now, 175 00:09:26,455 --> 00:09:31,179 you know, mathematically, I think of this like code and I think of it in functions. 176 00:09:31,179 --> 00:09:34,162 And I'm like, well, that just means that the answer is not a number, 177 00:09:34,162 --> 00:09:36,724 it's a function, and and we should be okay with that. 178 00:09:36,724 --> 00:09:41,668 But this helps put it into more of a of a an average that they can 179 00:09:41,729 --> 00:09:47,254 use and benchmark even if it's n not necessarily accurate. 180 00:09:47,254 --> 00:09:48,254 And 181 00:09:48,534 --> 00:09:52,105 They wouldn't see it the way you'd necessarily experience it, 182 00:09:52,105 --> 00:09:54,536 right? In in nineteen eighty three when they made this switch, 183 00:09:54,536 --> 00:09:56,586 home prices were crazy volatile. 184 00:09:56,586 --> 00:09:58,988 Like inflation was crazy volatile. 185 00:09:58,988 --> 00:10:05,270 ~ my my mother in law worked at the time in in home loan mortgaging 186 00:10:05,730 --> 00:10:06,730 and 187 00:10:06,871 --> 00:10:12,033 the there were there were double digit numbers on interest rates at the time. 188 00:10:12,033 --> 00:10:15,444 And so they're like she has old forms that would have, 189 00:10:15,444 --> 00:10:18,225 you know, two two places for the interest rate. 190 00:10:18,807 --> 00:10:23,829 ~ whereas you know in our adulthood coming up, 191 00:10:23,829 --> 00:10:28,381 we've never really seen double digit interest rates on mortgages, 192 00:10:28,381 --> 00:10:30,011 especially. ~ 193 00:10:29,890 --> 00:10:30,890 Mm-hmm. 194 00:10:30,752 --> 00:10:33,452 it's usually been confined to single digit rates. 195 00:10:33,452 --> 00:10:38,194 It's something to seven to eight percent was the average when I first entered into 196 00:10:38,194 --> 00:10:43,691 the housing market. ~ but you know that's that's neither here or there. 197 00:10:43,691 --> 00:10:46,946 I I kind of get why they wanted something stable. 198 00:10:48,129 --> 00:10:52,251 But by switching from the prices to this constructed rent, 199 00:10:52,251 --> 00:10:56,432 they basically created a measurement that lags behind reality. 200 00:10:56,432 --> 00:10:59,874 So if if real housing costs go up, 201 00:10:59,874 --> 00:11:03,826 which if you own a home, the things that go up are your you know, 202 00:11:03,826 --> 00:11:05,656 your not your mortgage, your mortgage is fixed, 203 00:11:05,656 --> 00:11:10,268 but your insurance goes up and your your insurance goes up and your taxes 204 00:11:10,268 --> 00:11:13,960 go up and and all of those things mean that you're actually paying more month 205 00:11:13,960 --> 00:11:17,031 to month, regardless of the fact that your mortgage isn't changing. 206 00:11:17,953 --> 00:11:19,232 But if 207 00:11:19,732 --> 00:11:22,993 Based on surveys and like lagging data and models, 208 00:11:22,993 --> 00:11:26,134 that's what actually gets to the important part, 209 00:11:26,134 --> 00:11:27,594 right? Like housing got expensive. 210 00:11:27,594 --> 00:11:30,935 Very expensive, and it continues to get higher month to month. 211 00:11:30,935 --> 00:11:34,956 But the CPI shelter component keeps kind 212 00:11:34,956 --> 00:11:38,397 of roughing over, 213 00:11:38,397 --> 00:11:44,718 sandpaper, if you will, ~ that reality with a number that nobody actually pays, 214 00:11:44,718 --> 00:11:47,619 it disconnects itself from reality. 215 00:11:48,119 --> 00:11:50,901 And that's been the status quo for roughly forty years. 216 00:11:51,521 --> 00:11:55,143 ~ gross. ~ that because that was just one year before I was born. 217 00:11:55,143 --> 00:12:00,345 Well, yes. ~ for forty years the the main measure of inflation that the 218 00:12:00,345 --> 00:12:05,477 Fed and the public both use has a built in lag 219 00:12:06,017 --> 00:12:12,100 on the biggest expense most individual human beings have in the United States. 220 00:12:12,100 --> 00:12:16,576 And really no one talks about it because it's technical, 221 00:12:16,576 --> 00:12:17,862 right? It's buried 222 00:12:18,467 --> 00:12:21,025 In the methodology, in the formula, 223 00:12:21,025 --> 00:12:23,502 because the alternative sounds more elegant, 224 00:12:23,502 --> 00:12:26,241 but only in theory and only if you're good at math. 225 00:12:26,741 --> 00:12:29,824 Right, and as you mentioned, when you're living in the mathematical abstract world, 226 00:12:29,824 --> 00:12:32,187 this seems like a lot easier to compute, 227 00:12:32,187 --> 00:12:36,472 but when actual people are depending on the the numbers and what they imply, 228 00:12:36,472 --> 00:12:39,085 there's a cost, practical cost to this, 229 00:12:39,085 --> 00:12:40,085 right? 230 00:12:39,962 --> 00:12:40,962 Yes, 231 00:12:40,452 --> 00:12:43,955 that's exactly right. It costs something in the practice of it. 232 00:12:43,955 --> 00:12:50,829 Because shelter is now about thirty five percent of the the CPI basket, 233 00:12:50,829 --> 00:12:54,262 like the entirety of the consumer price index, 234 00:12:54,262 --> 00:12:56,443 thirty five percent of it, that's more than a third, 235 00:12:56,443 --> 00:13:01,897 is the single biggest weight of the average person's burden. 236 00:13:01,897 --> 00:13:08,131 And if that weight is systematically understating the actual costs of housing. 237 00:13:08,631 --> 00:13:11,937 Then the entire index is undershooting the inflation that people 238 00:13:11,937 --> 00:13:12,979 are actually feeling. 239 00:13:13,479 --> 00:13:15,833 Right, so this methodological choice they've made, 240 00:13:15,833 --> 00:13:18,858 ~ which is to measure something that doesn't actually exist, 241 00:13:18,858 --> 00:13:21,141 ~ now that we've had it for forty years, 242 00:13:21,141 --> 00:13:24,928 there must be some kind of consequences we can point to and see how 243 00:13:24,928 --> 00:13:27,612 it affected real people, for better or worse. 244 00:13:26,284 --> 00:13:27,594 Yeah, it i it it has 245 00:13:28,095 --> 00:13:30,415 for sure, it has had consequences. 246 00:13:30,415 --> 00:13:33,216 And they actually got worse in nineteen ninety six. 247 00:13:33,216 --> 00:13:36,647 ~ the Senate Finance Committee commissioned a study on whether 248 00:13:36,647 --> 00:13:39,978 the consumer price index was overstating inflation. 249 00:13:39,978 --> 00:13:42,839 It was called the Boskin Commission, 250 00:13:42,839 --> 00:13:45,219 named after the chair, Michael Boskin. 251 00:13:45,219 --> 00:13:48,560 Their their ultimate conclusion was pretty direct. 252 00:13:48,560 --> 00:13:54,152 The CPI overstates inflation by about one point one percentage points per year. 253 00:13:54,652 --> 00:13:57,806 It's not a tiny amount when you're talking about a an average of 254 00:13:57,806 --> 00:14:00,590 a three percent annual. Like that's more than a third. 255 00:14:00,590 --> 00:14:07,290 Again. ~ we're talking ~ about a third of of headline inflation just being 256 00:14:07,410 --> 00:14:08,822 incorrect, being wrong. 257 00:14:09,793 --> 00:14:11,436 And what did they base that conclusion on? 258 00:14:11,936 --> 00:14:14,527 They looked at the at substitution bias. 259 00:14:14,527 --> 00:14:17,168 It's the idea that when prices go up, 260 00:14:17,168 --> 00:14:20,049 people buy different things, cheaper things. 261 00:14:20,049 --> 00:14:23,450 And the the index doesn't fully account for any of that. 262 00:14:23,450 --> 00:14:26,902 They looked at the hedonic quality adjustments. 263 00:14:26,902 --> 00:14:29,653 The idea that when a product gets better, 264 00:14:29,653 --> 00:14:34,134 its price increases isn't really an increase because you're getting something 265 00:14:34,134 --> 00:14:36,855 better. ~ to kind of restate that, 266 00:14:36,855 --> 00:14:39,476 they looked at at outlet substitution. 267 00:14:39,476 --> 00:14:41,237 This idea that people switch 268 00:14:41,619 --> 00:14:45,361 where they shop. They go from Target to Walmart. 269 00:14:45,361 --> 00:14:50,794 They go from ~ JCPenney to TJ Maxx, 270 00:14:50,794 --> 00:14:56,437 right? Like they all of these things are defensible in theory. 271 00:14:56,437 --> 00:15:01,729 The problem is that the commission used these ideas to justify new methodologies, 272 00:15:01,729 --> 00:15:06,662 new calculations, new ways of coming to these numbers. 273 00:15:06,662 --> 00:15:10,093 And those methodologies, once they were adopted, 274 00:15:10,093 --> 00:15:11,164 had a direct 275 00:15:11,444 --> 00:15:12,765 Policy consequence. 276 00:15:13,265 --> 00:15:15,975 I could guess that whether that's a good or bad consequence, 277 00:15:15,975 --> 00:15:17,260 but I'll just let you tell us. 278 00:15:17,260 --> 00:15:18,385 Which was it? 279 00:15:17,747 --> 00:15:18,747 Yeah, I mean it 280 00:15:18,578 --> 00:15:20,839 it's almost never good for the average person, 281 00:15:20,839 --> 00:15:21,839 right? 282 00:15:21,468 --> 00:15:22,468 Right. 283 00:15:22,179 --> 00:15:24,941 Co cost of living adjustments for Social Security, 284 00:15:24,941 --> 00:15:29,963 colas, the way which is what the way cost of living adjustments, 285 00:15:29,963 --> 00:15:33,105 the way that social security payments adjust every year, 286 00:15:33,105 --> 00:15:36,227 is directly tied to the consumer price index. 287 00:15:36,227 --> 00:15:40,749 So if the consumer price index is wrong by one point one percentage points, 288 00:15:40,749 --> 00:15:45,662 then Social Security is paying out one point one percentage points more. 289 00:15:46,313 --> 00:15:48,123 Than the government thinks that it should. 290 00:15:48,123 --> 00:15:51,294 It's almost never that direction. 291 00:15:51,294 --> 00:15:53,375 It's usually it's paying out less 292 00:15:52,573 --> 00:15:53,573 Right. 293 00:15:53,915 --> 00:15:58,576 than people actually need. So if the Boschian Commission was right that 294 00:15:58,576 --> 00:16:03,557 the CPI was overstating inflation by 1.1 percentage points, 295 00:16:03,557 --> 00:16:09,779 then reducing that overstatement means reducing Social Secures. 296 00:16:09,779 --> 00:16:15,020 Which is a nice way of saying we re have to reduce. 297 00:16:15,326 --> 00:16:20,230 The Social Security payments that people have grown accustomed to receiving. 298 00:16:20,230 --> 00:16:24,895 So the Commission's own projections basically said the changes they recommended 299 00:16:24,895 --> 00:16:31,073 would cut $148 billion in deficit spending and six hundred 300 00:16:31,073 --> 00:16:32,712 and ninety-one billion dollars, 301 00:16:32,712 --> 00:16:36,775 blah blah blah billion, off of the national debt by two thousand six. 302 00:16:36,775 --> 00:16:40,839 Now that's real money. Those are real people's benefits. 303 00:16:41,339 --> 00:16:44,224 So this measurement change that they came up with now has had 304 00:16:44,224 --> 00:16:45,936 a direct physical consequence, 305 00:16:45,936 --> 00:16:48,611 basically. Yeah. 306 00:16:46,340 --> 00:16:48,331 And it still it still does, directly. 307 00:16:48,331 --> 00:16:49,331 Like 308 00:16:48,622 --> 00:16:53,165 it was not an abstract debate about how it was done. 309 00:16:53,165 --> 00:16:57,708 And this was a commissioned study that concluded the government 310 00:16:57,708 --> 00:17:01,770 was overpaying retirees because the inflation measurement was wrong. 311 00:17:01,770 --> 00:17:07,014 And their conclusion led to policy changes that reduced the amount that retirees 312 00:17:07,014 --> 00:17:10,616 got. Now some of those changes were reasonable. 313 00:17:10,616 --> 00:17:14,529 Don't cut my head off here. But substitution bias is real. 314 00:17:15,129 --> 00:17:17,061 You do change what you buy, right? 315 00:17:17,061 --> 00:17:22,725 You don't buy if if you know ground beef goes up like it's been going 316 00:17:22,725 --> 00:17:25,577 up lately to nine dollars a pound, 317 00:17:25,577 --> 00:17:27,429 you're gonna switch to ground pork. 318 00:17:27,429 --> 00:17:32,142 That's still currently sitting at about what beef was before Donald Trump 319 00:17:32,142 --> 00:17:35,594 was elected. So maybe ~ I don't know, 320 00:17:35,594 --> 00:17:38,516 ~ inside the commission, go ahead. 321 00:17:36,984 --> 00:17:37,984 But 322 00:17:38,937 --> 00:17:41,291 sorry I was gonna say I mean these that sounds reasonable, 323 00:17:41,291 --> 00:17:43,985 but then these numbers are just kinda like living in people's heads. 324 00:17:43,985 --> 00:17:47,450 There's no way unless you can interview everybody at every checkout line 325 00:17:47,450 --> 00:17:48,953 to figure out, you know, why are you buying pork? 326 00:17:48,953 --> 00:17:51,677 Did you switch from from from beef to pork because of 327 00:17:52,482 --> 00:17:54,643 The costs or health reasons or because, 328 00:17:54,643 --> 00:17:58,714 you know, whatever. You you you left your Judais Judaism faith. 329 00:17:58,714 --> 00:18:00,425 I mean whatever. ~ 330 00:17:59,292 --> 00:18:00,292 Right. 331 00:18:00,425 --> 00:18:02,756 you know, like what what's going on here? 332 00:18:02,756 --> 00:18:04,737 And we have to imagine that we can summarize that, 333 00:18:04,737 --> 00:18:07,978 but it just seems like it's depending on a lot of stuff that's not really there's 334 00:18:07,978 --> 00:18:09,709 no way to verify or refute, right? 335 00:18:10,037 --> 00:18:11,037 It is. I 336 00:18:10,968 --> 00:18:16,553 it's absolutely doing that. And it's not something that is truly data if 337 00:18:16,553 --> 00:18:20,176 it r relies upon volunteers submitting information willingly. 338 00:18:20,176 --> 00:18:24,230 Right? Like you might not actually think about the fact that you're switching 339 00:18:24,230 --> 00:18:25,972 to pork, but when you're 340 00:18:25,655 --> 00:18:26,655 Right. 341 00:18:25,972 --> 00:18:28,255 sitting there looking at the price tag and you're going, 342 00:18:28,255 --> 00:18:34,269 Why is it like eleven dollars more than it was three months ago for the same amount? 343 00:18:34,561 --> 00:18:38,015 I'm gonna go to the one that's closer to the price because we we fixate 344 00:18:38,015 --> 00:18:40,147 on on the actual amount that it costs. 345 00:18:40,147 --> 00:18:41,788 Rather than 346 00:18:41,346 --> 00:18:42,346 Hm. 347 00:18:43,310 --> 00:18:45,112 the reasoning behind why we're doing it. 348 00:18:45,112 --> 00:18:47,504 It happens so innately that we don't really think about it. 349 00:18:47,504 --> 00:18:49,146 We're like, well, I mean it's it's meat, 350 00:18:49,146 --> 00:18:50,877 it's right next to it. I'm not Jewish. 351 00:18:50,877 --> 00:18:53,831 Like, you know, like I I'll substitute that. 352 00:18:54,331 --> 00:18:59,803 Yeah. So even inside the commission there w they were kind of upset about this. 353 00:18:59,803 --> 00:19:01,323 They was they were disagreeing about it. 354 00:19:01,323 --> 00:19:05,485 Robert Gordon, who was the ~ just one of the members of this commission, 355 00:19:05,485 --> 00:19:08,216 said that the bias estimate should have been higher, 356 00:19:08,216 --> 00:19:13,248 not lower. Meaning maybe one point two to one point three percent, 357 00:19:13,248 --> 00:19:18,230 not one point one. And then Dean Baker at the Economic Policy Institute, 358 00:19:18,230 --> 00:19:21,851 he looked at the evidence and said it was largely speculative. 359 00:19:22,015 --> 00:19:23,781 Guesswork rather than careful research, 360 00:19:23,781 --> 00:19:26,842 and that the Commission had failed to make a compelling case that 361 00:19:26,842 --> 00:19:28,719 the adjustments were justified. 362 00:19:29,219 --> 00:19:31,775 So they did a study to justify policy outcome? 363 00:19:32,275 --> 00:19:38,426 ~ yeah, kinda. another reason is that they genuinely believed and and 364 00:19:38,426 --> 00:19:41,867 I I do believe this, I think they believed that the consumer price index 365 00:19:41,867 --> 00:19:48,839 was overestimating inflation. But they were very much less careful about 366 00:19:48,839 --> 00:19:50,559 the evidence than they probably should have been. 367 00:19:50,559 --> 00:19:53,340 They didn't use really it necessarily the scientific method. 368 00:19:53,340 --> 00:19:55,861 But either way, the outcome was the same. 369 00:19:55,861 --> 00:20:01,002 They changed the way that it was calculated in a way that conveniently reduced 370 00:20:01,536 --> 00:20:03,187 What the government had to pay out. 371 00:20:03,187 --> 00:20:07,248 And that change is now baked into how we measure inflation. 372 00:20:07,248 --> 00:20:09,540 And that was nineteen ninety six, 373 00:20:09,540 --> 00:20:10,970 that's almost thirty years ago. 374 00:20:10,970 --> 00:20:15,182 And we're still living with the consequences of that choice. 375 00:20:15,682 --> 00:20:18,986 So you've got the nineteen eighty three decision to use something that nobody pays 376 00:20:18,986 --> 00:20:23,111 for, and then the nineteen ninety-six decision to adjust the methodology 377 00:20:23,111 --> 00:20:25,573 to suggest that inflation is lower than people think it is. 378 00:20:26,547 --> 00:20:28,348 Right. And that that was just then. 379 00:20:28,348 --> 00:20:32,491 We'll get a little bit in into what's happened over the past, 380 00:20:32,491 --> 00:20:33,872 you know, year and a half or so. 381 00:20:33,872 --> 00:20:37,014 Two almost two. I don't hard to tell at this point. 382 00:20:37,014 --> 00:20:38,735 Two deliberate choices, right? 383 00:20:38,735 --> 00:20:40,146 One was building on the other. 384 00:20:40,146 --> 00:20:43,018 Both with technical justifications, 385 00:20:43,018 --> 00:20:49,403 both had the practical effect of making the official inflation number lower than 386 00:20:49,883 --> 00:20:51,944 lived and experienced inflation. 387 00:20:51,944 --> 00:20:54,076 But both of them were sitting too much. 388 00:20:54,449 --> 00:20:59,482 Inside the structure where measurement sounds neutral and scientific, 389 00:20:59,482 --> 00:21:00,923 like a tape measure. You know, 390 00:21:00,923 --> 00:21:03,035 you you didn't say, Hey, go measure that symbol, 391 00:21:03,035 --> 00:21:06,637 right, over there and and tell me how wide it is, 392 00:21:06,637 --> 00:21:07,637 and you look at it and you go, 393 00:21:07,488 --> 00:21:10,220 Well, it's eight inches. It that feels objective. 394 00:21:10,220 --> 00:21:14,806 But when you have to measure it in millimeters and you tell me it's eight inches, 395 00:21:14,806 --> 00:21:17,825 I can convert that. But if you need a whole number, 396 00:21:17,825 --> 00:21:19,726 you know, you want to know how many inches it was, 397 00:21:19,726 --> 00:21:22,518 it it doesn't really work out that way. 398 00:21:23,470 --> 00:21:28,077 But the policy consequence is that people who were on a fixed income, 399 00:21:28,077 --> 00:21:31,932 like retirees, but also people who are just working everyday jobs, 400 00:21:31,932 --> 00:21:35,968 whose income is tied to inflation adjustments, 401 00:21:35,968 --> 00:21:37,270 they just get less money. 402 00:21:37,770 --> 00:21:40,121 So in December of twenty twelve, 403 00:21:40,121 --> 00:21:44,102 the Federal Open Market Committee made another decision. 404 00:21:44,102 --> 00:21:50,123 They officially adopted the personal consumption expenditure price index. 405 00:21:50,123 --> 00:21:52,624 And look, the this is term heavy. 406 00:21:52,624 --> 00:21:54,364 This is definitely a term heavy episode. 407 00:21:54,364 --> 00:21:59,355 PCE is what they called it as their preferred measure for the 408 00:21:59,355 --> 00:22:02,166 two percent inflation target, not CPI. 409 00:22:02,166 --> 00:22:05,027 PCE. The one to 410 00:22:05,527 --> 00:22:08,839 The one on one hears about on the news, 411 00:22:08,839 --> 00:22:14,022 right? So the the word you actually hear about on the news is CPI, 412 00:22:14,022 --> 00:22:18,025 PCE. ~ You have to figure it out. 413 00:22:18,525 --> 00:22:20,018 So what's the the difference? 414 00:22:20,518 --> 00:22:23,959 So the PCE and CPI basically measure the same thing, 415 00:22:23,959 --> 00:22:26,950 right? Inflation. And that's what they're trying to do. 416 00:22:26,950 --> 00:22:28,870 But they do it in different ways. 417 00:22:28,870 --> 00:22:31,571 They they actually obtain the information and calculate it differently. 418 00:22:31,571 --> 00:22:34,332 It's a different basket of goods, 419 00:22:34,332 --> 00:22:38,093 different weightings, different calculations. 420 00:22:38,093 --> 00:22:42,034 The Fed says that PCE is more comprehensive, 421 00:22:42,034 --> 00:22:47,556 that it captures the actual pattern of what people spend money on more accurately. 422 00:22:47,556 --> 00:22:48,836 And look. 423 00:22:49,210 --> 00:22:52,563 There are reasonable technical arguments in that direction. 424 00:22:52,563 --> 00:22:58,277 But what actually matters is that PCE is consistently lower than 425 00:22:58,277 --> 00:23:01,289 the consumer price index, and not by a tiny amount. 426 00:23:01,289 --> 00:23:06,574 By enough that we're at two percent of inflation sounds a lot different from we're 427 00:23:06,574 --> 00:23:08,395 at two point five percent inflation. 428 00:23:08,895 --> 00:23:11,443 So the Fed is targeting a different number than what people hear. 429 00:23:11,943 --> 00:23:16,694 Yes, mostly, but they're targeting a different number than what people hear 430 00:23:16,694 --> 00:23:22,907 in a way that changes the public debate about inflation and what it's anchored 431 00:23:22,907 --> 00:23:26,998 on to the number that the Fed isn't actually using anymore. 432 00:23:26,998 --> 00:23:33,181 So imagine if the Fed were keeping ~ the economy at two percent unemployment, 433 00:23:33,181 --> 00:23:38,753 okay? And then they measured unemployment using a completely different calculation. 434 00:23:39,179 --> 00:23:43,004 That only counted employed people who also worked overtime. 435 00:23:43,004 --> 00:23:46,088 And then they just never explained that to anybody. 436 00:23:46,088 --> 00:23:50,685 And then they made policy based on the second number while Congress 437 00:23:50,685 --> 00:23:52,457 and the news talked about the first number. 438 00:23:52,457 --> 00:23:53,979 That's what's happened. 439 00:23:54,479 --> 00:23:58,759 And in that situation people would think they're insane or going insane because 440 00:23:58,759 --> 00:24:01,335 the official number and their actual lives don't match up anymore. 441 00:24:02,273 --> 00:24:05,005 Yeah, and that's a that's a lot of what happened at the ~ toward the 442 00:24:05,005 --> 00:24:08,648 end of of Biden's term in the same kind of concept. 443 00:24:08,648 --> 00:24:10,029 You know, they they kept telling us like, 444 00:24:10,029 --> 00:24:12,161 no, everything's improving, the numbers are getting better, 445 00:24:12,161 --> 00:24:15,554 everything's cheaper. And when we actually went to the grocery store, 446 00:24:15,554 --> 00:24:19,687 when we actually went places to do things and get the very things that 447 00:24:19,687 --> 00:24:22,860 we need to live and survive, it didn't really add up. 448 00:24:22,860 --> 00:24:27,003 They're not going insane necessarily, 449 00:24:27,003 --> 00:24:30,956 d just the numbers don't match up because they're they're literally measuring 450 00:24:30,956 --> 00:24:31,956 different things. 451 00:24:32,187 --> 00:24:38,969 So today, the the twelfth, the Bureau of Labor Statistics released the July CPI. 452 00:24:38,969 --> 00:24:43,010 And the headline is three point four percent with the core at 453 00:24:43,010 --> 00:24:47,231 two point five percent. But the Fed's looking at that price pricing consumer 454 00:24:47,811 --> 00:24:51,152 equivalent, which probably came in lower. 455 00:24:51,152 --> 00:24:53,232 But that's not gonna make a headline. 456 00:24:53,232 --> 00:24:56,383 And that's not a detail at all. 457 00:24:56,383 --> 00:24:59,534 That's kind of the whole reason the Fed decided not to move rates 458 00:24:59,914 --> 00:25:01,094 on july twenty ninth. 459 00:25:01,690 --> 00:25:04,910 And held rates at three point five to three point seven five percent. 460 00:25:05,410 --> 00:25:07,209 And if memory serves that was a tight vote, 461 00:25:07,209 --> 00:25:08,209 wasn't it? 462 00:25:08,094 --> 00:25:11,626 Nine nine to three, I mean three members dissented. 463 00:25:11,626 --> 00:25:17,549 They wanted higher rates. The dissent isn't usually that visible, 464 00:25:17,549 --> 00:25:23,312 but it was this time. And I know you well enough to know that your next question 465 00:25:23,312 --> 00:25:26,564 is going to be why? Because even within the Fed, 466 00:25:26,564 --> 00:25:29,496 there's a lot of disagreement about whether that number, 467 00:25:29,496 --> 00:25:31,957 right, the PCE number that they're actually looking at, 468 00:25:31,957 --> 00:25:35,849 is actually telling them what they need to know about inflation right now. 469 00:25:36,531 --> 00:25:41,437 Is inflation really cooling or are the metrics they've chosen just 470 00:25:41,477 --> 00:25:43,399 not seeing something that's really happening? 471 00:25:43,899 --> 00:25:46,704 And meanwhile the CPI number, the one everyone's talking about, 472 00:25:46,704 --> 00:25:50,089 is showing shelter up three point four percent year over year, 473 00:25:50,089 --> 00:25:53,277 and it drove nearly two third of two thirds of the monthly increase. 474 00:25:53,777 --> 00:25:55,709 Exactly. It it drove two thirds. 475 00:25:55,709 --> 00:25:58,952 So when people see three point four, 476 00:25:58,952 --> 00:26:01,734 they might think, okay, inflation's spread around, 477 00:26:01,734 --> 00:26:07,900 but it's not. It's concentrated in the one cost that none of us can actually escape. 478 00:26:07,900 --> 00:26:12,835 And that concentration showing up in the number that the public's debating about 479 00:26:12,835 --> 00:26:14,947 is not the number the Fed's using to make decisions. 480 00:26:15,447 --> 00:26:19,199 So why does shelters stay that elevated if the housing market's cooled off? 481 00:26:19,699 --> 00:26:22,782 Now we're gonna introduce another three letter acronym. 482 00:26:22,782 --> 00:26:24,983 That's because the OER is lagging. 483 00:26:23,829 --> 00:26:26,782 Okay. That's what we're missing. 484 00:26:24,983 --> 00:26:26,999 So it's based on 485 00:26:27,205 --> 00:26:31,989 right. O OER, of course. So it's based on survey data. 486 00:26:31,989 --> 00:26:35,352 Yes, actual housing sales have cooled, 487 00:26:35,352 --> 00:26:40,496 but the surveys are slow to catch up because rental surveys take time. 488 00:26:40,496 --> 00:26:45,500 The data looks back. So you get this ghost of 489 00:26:45,720 --> 00:26:47,862 of housing's past inflation. 490 00:26:48,300 --> 00:26:52,192 kind of baked into the index even after the markets started to move. 491 00:26:52,192 --> 00:26:54,834 And the Fed is looking at PCE, 492 00:26:54,834 --> 00:26:58,756 which probably doesn't weight shelter as heavily. 493 00:26:58,756 --> 00:27:04,269 So they see it cooling. The num the public sees the CPI headline and thinks it's 494 00:27:04,269 --> 00:27:08,572 not cooling. And both are looking at their number and thinking that the other 495 00:27:08,572 --> 00:27:09,572 one is crazy. 496 00:27:09,692 --> 00:27:12,390 And then there's the actual problem of data collection breaking down. 497 00:27:12,937 --> 00:27:16,030 Yeah. So in October of twenty twenty five, 498 00:27:16,030 --> 00:27:20,213 the government shut down. Not metaphorically it shut down, 499 00:27:20,213 --> 00:27:21,894 and the Bureau of Labor Statistics, 500 00:27:21,894 --> 00:27:25,818 which produces the CPI, is a government agency. 501 00:27:25,818 --> 00:27:29,941 Their staff were furloughed, they stopped collecting data, 502 00:27:29,941 --> 00:27:33,035 and when they came back and tried to put it together, 503 00:27:33,035 --> 00:27:36,277 the shelter component of the CPI or the housing, 504 00:27:36,277 --> 00:27:37,648 maybe we should just call it housing, 505 00:27:37,648 --> 00:27:40,210 but the housing component of the CPI for that period. 506 00:27:40,722 --> 00:27:43,003 Had a massive data gap. There just wasn't any. 507 00:27:43,003 --> 00:27:45,824 So they couldn't actually survey people. 508 00:27:45,824 --> 00:27:48,224 They were not literally allowed to show up to work. 509 00:27:48,224 --> 00:27:50,435 So they couldn't get new rental information. 510 00:27:50,435 --> 00:27:57,006 So they made a decision. They carried forward the previous month's housing estimate. 511 00:27:57,006 --> 00:27:59,328 And then they froze it in place, 512 00:27:59,328 --> 00:28:05,860 which assumed a 0% change in housing inflation for that shutdown period. 513 00:28:06,360 --> 00:28:10,287 So they just filled the hole in the data by keeping the number the same 514 00:28:10,287 --> 00:28:12,832 for the previous month or previous ~ report. 515 00:28:13,303 --> 00:28:15,924 Exactly. And that's not great. 516 00:28:15,924 --> 00:28:17,324 So Jason 517 00:28:16,887 --> 00:28:17,887 No. 518 00:28:17,404 --> 00:28:21,425 Furman, who's at Harvard now, he actually used to chair the Council 519 00:28:21,425 --> 00:28:24,466 on Economic Advisors under President Obama. 520 00:28:24,466 --> 00:28:27,777 He looked at that and called it a big judgment error. 521 00:28:27,777 --> 00:28:33,268 I mean, I would too. But he said the the carry forward methodology probably 522 00:28:33,268 --> 00:28:39,149 understated housing inflation by point three to six percentage points year over year 523 00:28:39,149 --> 00:28:41,610 during that period of time. That's not trivial. 524 00:28:42,110 --> 00:28:45,285 Because housing is a third of the consumer price index. 525 00:28:45,285 --> 00:28:49,431 So if if shelter if shelter or housing is undercounted by a half 526 00:28:49,431 --> 00:28:52,806 of a percentage point, that spreads through the whole index. 527 00:28:53,306 --> 00:28:55,459 Which ~ breaks the measurement system itself. 528 00:28:55,959 --> 00:28:58,340 It's broke. And kind of more than that. 529 00:28:58,340 --> 00:29:00,130 When the government shuts down, 530 00:29:00,130 --> 00:29:04,142 you get this weird waterfall where the data stops, 531 00:29:04,142 --> 00:29:08,343 the analysis stops, and then you've got to fill in the gaps with assumptions. 532 00:29:08,343 --> 00:29:12,404 And those assumptions are baked into the official number that people 533 00:29:12,404 --> 00:29:15,905 see and that the Fed is supposed to be using. 534 00:29:15,905 --> 00:29:18,566 Except the Fed is using the PCE, 535 00:29:18,566 --> 00:29:20,867 which has its own separate calculation. 536 00:29:20,867 --> 00:29:21,927 So 537 00:29:22,427 --> 00:29:25,820 Who knows whether their number got affected the same way or not? 538 00:29:27,204 --> 00:29:28,204 This is insane. 539 00:29:29,080 --> 00:29:34,371 It is. It's a measurement system that's pretty much already laggy, 540 00:29:34,371 --> 00:29:37,112 right? Like it's already using hypothetical numbers instead of the ones that we're 541 00:29:37,112 --> 00:29:39,673 seeing right now, because that's just how data works. 542 00:29:39,673 --> 00:29:43,074 It has to be collected. So now it's also vulnerable 543 00:29:43,074 --> 00:29:47,025 to government shutdowns literally breaking the pipeline of information. 544 00:29:47,025 --> 00:29:50,776 It's not like, you know, it it's just a log that pushes into 545 00:29:50,790 --> 00:29:54,587 a database somewhere and and when people aren't at work it just keeps pushing. 546 00:29:55,087 --> 00:29:57,378 It it just breaks and there's just no information. 547 00:29:57,378 --> 00:30:02,210 And the only reason people even know about this is because Furman actually published 548 00:30:02,210 --> 00:30:05,612 a note about it. It's not in the official CPI release, 549 00:30:05,612 --> 00:30:08,373 it's not flagged, there's no blaring alarm. 550 00:30:08,373 --> 00:30:12,135 You just have to go digging to find out the the consumer price index you're looking 551 00:30:12,135 --> 00:30:16,927 at has a little star next to it saying that the housing component 552 00:30:16,927 --> 00:30:20,259 was estimated because the government didn't exist during that period of time. 553 00:30:20,759 --> 00:30:24,492 So let's talk about what we need to know at this point. 554 00:30:24,492 --> 00:30:29,515 what I want to know is you go to the grocery store, 555 00:30:29,515 --> 00:30:31,657 you know what you spent last summer, 556 00:30:31,657 --> 00:30:33,598 you know what you spent last month, 557 00:30:33,598 --> 00:30:36,850 and the government tells you food inflation is two point seven percent year over 558 00:30:36,850 --> 00:30:37,850 year. 559 00:30:37,581 --> 00:30:40,548 And you know that that's not what you're actually paying for at the register. 560 00:30:41,341 --> 00:30:43,124 Right, I know it. Everyone knows it. 561 00:30:43,124 --> 00:30:45,258 So where is the two point seven percent coming from? 562 00:30:45,884 --> 00:30:47,285 Coming from a different calculator, 563 00:30:47,285 --> 00:30:49,606 a different sample, a different basket. 564 00:30:49,606 --> 00:30:53,588 It's it's food at home inflation specifically, 565 00:30:53,588 --> 00:30:58,851 which the Bureau of Labor Statistics measures using a specific approach. 566 00:30:58,851 --> 00:31:00,852 But researchers have looked at this. 567 00:31:00,852 --> 00:31:05,915 They've asked consumers how much they think grocery prices have gone up. 568 00:31:05,915 --> 00:31:11,458 Now, Pew Research found that the median consumer thinks grocery inflation 569 00:31:11,458 --> 00:31:13,640 is somewhere between 3.9%. 570 00:31:14,566 --> 00:31:17,718 and five point four percent, not two point seven. 571 00:31:17,718 --> 00:31:23,240 And that gap is consistent in perpet perpetuity for all of time. 572 00:31:23,240 --> 00:31:28,943 It keeps showing up. People think inflation is higher than the actual number says 573 00:31:28,943 --> 00:31:29,943 it is. 574 00:31:29,763 --> 00:31:32,131 And that's a two to three percent percentage point gap. 575 00:31:32,631 --> 00:31:34,372 On the low end of what people report. 576 00:31:34,372 --> 00:31:35,823 And here's the thing about that number. 577 00:31:35,823 --> 00:31:38,244 It's not that people are bad at math. 578 00:31:38,244 --> 00:31:42,166 A lot of people are. It's that the official number is using 579 00:31:42,321 --> 00:31:47,348 a statistics approach that doesn't actually match how people actually shop. 580 00:31:47,348 --> 00:31:50,870 The CPI uses a fixed basket of items. 581 00:31:50,870 --> 00:31:53,502 Your spending patterns change, 582 00:31:53,502 --> 00:31:55,423 right? Like you stop buying the expensive thing, 583 00:31:55,423 --> 00:31:56,753 you buy the store brand instead, 584 00:31:56,753 --> 00:32:00,435 you buy less. The index, say butter went up 20%, 585 00:32:00,435 --> 00:32:02,036 but you switch to a different 586 00:32:02,190 --> 00:32:04,131 product that only went up ten percent. 587 00:32:04,131 --> 00:32:08,144 So the index says your cost went up less than it actually did. 588 00:32:08,144 --> 00:32:14,648 Your lived experience is you're paying more for less like actual item, 589 00:32:14,648 --> 00:32:18,801 but the measurement smooths out over the substitution effect. 590 00:32:19,301 --> 00:32:21,644 The metric is built to absorb your adaptation. 591 00:32:22,265 --> 00:32:24,066 Right. And then on top of that, 592 00:32:24,066 --> 00:32:29,040 the BLS measures food at home completely separate as from food away from home. 593 00:32:29,040 --> 00:32:32,792 So if you're eating at a restaurant or a takeout, 594 00:32:32,792 --> 00:32:34,484 that's in a completely different bucket, 595 00:32:34,484 --> 00:32:35,805 completely different category. 596 00:32:35,805 --> 00:32:37,886 When you're actually budgeting, 597 00:32:37,886 --> 00:32:40,078 you're thinking about feeding yourself. 598 00:32:40,078 --> 00:32:42,890 Like that's it. You're like food in my belle. 599 00:32:42,890 --> 00:32:46,152 The index is splitting three ways though. 600 00:32:46,152 --> 00:32:48,064 And the result is that housing 601 00:32:48,417 --> 00:32:49,868 which is thirty five percent of that, 602 00:32:49,868 --> 00:32:52,860 is measured with a phantom rent number no one actually pays. 603 00:32:52,860 --> 00:32:56,603 And food, which is actually hitting people the hardest, 604 00:32:56,603 --> 00:33:01,147 is measured with a calculation that assumes that you never change your consumption 605 00:33:01,147 --> 00:33:02,598 pattern and you never eat out. 606 00:33:03,119 --> 00:33:05,240 official grocery number since twenty fifteen, 607 00:33:05,240 --> 00:33:07,252 and think about that time window, 608 00:33:07,252 --> 00:33:08,763 that's when smartphones became, 609 00:33:08,763 --> 00:33:13,987 you know, necessary and when housing got really expensive in most cities, 610 00:33:13,987 --> 00:33:15,828 when everything kind of tightened up, 611 00:33:15,828 --> 00:33:21,473 grocery prices are up thirty six point nine percent since twenty fifteen. 612 00:33:21,473 --> 00:33:25,596 Since twenty twenty, they're up twenty five point two percent. 613 00:33:26,096 --> 00:33:28,158 That's right at that inflation time, 614 00:33:28,158 --> 00:33:33,402 that pandemic time, and the official CPI food number is two point seven percent. 615 00:33:33,402 --> 00:33:37,585 That disconnect is not small. That's not like a measurement error within 616 00:33:37,585 --> 00:33:40,647 a reasonable window. That is a structural error. 617 00:33:41,147 --> 00:33:43,475 And the person who goes to the store every week is experiencing 618 00:33:43,475 --> 00:33:47,067 the twenty five percent and not experiencing the two point seven percent. 619 00:33:47,465 --> 00:33:49,517 No, they're not. And that's not hypothetical. 620 00:33:49,517 --> 00:33:53,920 Like there's a federal employee named Chris Myers in Kansas City. 621 00:33:53,920 --> 00:33:58,803 He's a federal employee that used to co own crane brewing. 622 00:33:58,803 --> 00:34:01,956 During the October shutdown, he lost his paycheck, 623 00:34:01,956 --> 00:34:04,978 no income. His family was already on a tight budget, 624 00:34:04,978 --> 00:34:09,271 but once that shutdown hit, his wife told the Missouri Independent that their 625 00:34:09,271 --> 00:34:11,233 grocery budget nearly doubled. 626 00:34:11,233 --> 00:34:15,336 They went from spending between two hundred and three hundred dollars every 627 00:34:15,336 --> 00:34:16,437 two weeks on groceries. 628 00:34:16,801 --> 00:34:19,306 To four to five hundred dollars every two weeks. 629 00:34:19,306 --> 00:34:23,484 That's not the CPI number. That's what actually happened when 630 00:34:23,484 --> 00:34:27,302 an actual household with actual income got disrupted and they had 631 00:34:27,302 --> 00:34:28,935 to feed their actual children. 632 00:34:29,435 --> 00:34:30,435 When the government shut down. 633 00:34:30,916 --> 00:34:35,910 Yeah, th the government shut down and the same shutdown broke that data that 634 00:34:35,910 --> 00:34:40,234 we had for housing, so the measurement system was literally broken at 635 00:34:40,234 --> 00:34:44,387 the the one time when real families were experiencing real cost increases. 636 00:34:44,387 --> 00:34:47,240 There's also a woman named Megan Overfeldt. 637 00:34:47,240 --> 00:34:50,572 She's a data entry worker in her thirties and lives in Raytown, 638 00:34:50,572 --> 00:34:54,336 Missouri. She has four kids. She grew up in Johnson County, 639 00:34:54,336 --> 00:34:59,010 Kansas, and her childhood home now sells for three hundred and seventy five thousand 640 00:34:59,010 --> 00:35:00,010 dollars. 641 00:35:00,152 --> 00:35:02,595 She cannot afford to buy in the neighborhood she grew up in. 642 00:35:02,595 --> 00:35:05,699 She has a job, she's an employed adult, 643 00:35:05,699 --> 00:35:09,524 and the neighborhood of her childhood is priced well beyond her reach. 644 00:35:10,024 --> 00:35:12,807 Despite the official numbers saying housing inflation is easy. 645 00:35:13,307 --> 00:35:16,889 Right. And here's the thing the reason sh shelter, 646 00:35:16,889 --> 00:35:21,621 housing in the CPI is so low relative to what people are actually experiencing 647 00:35:22,061 --> 00:35:24,913 is that it's measured with owner's equivalent rent, 648 00:35:24,913 --> 00:35:29,404 the OER. It's asking people what they think they would pay to rent their 649 00:35:29,404 --> 00:35:33,547 own house to themselves. But what Megan Overfeld is actually confronting 650 00:35:33,547 --> 00:35:36,628 is sh can she afford to buy a house? 651 00:35:36,628 --> 00:35:38,439 Can she afford the down payment, 652 00:35:38,439 --> 00:35:40,650 the mortgage, the property taxes? 653 00:35:40,650 --> 00:35:42,161 Those are real costs. 654 00:35:43,005 --> 00:35:47,298 Paul Donovan at UBS Global Wealth Management looked at the CPI and said 655 00:35:47,298 --> 00:35:52,662 it pretty flat out. He said a quarter of the CPI basket is a complete fantasy. 656 00:35:52,662 --> 00:35:56,985 Owner's equivalent rent is a price that no one actually pays. 657 00:35:56,985 --> 00:36:01,888 And he calculated that by using the OER instead of actual housing costs, 658 00:36:01,888 --> 00:36:07,613 the Federal Reserve has systematically understated inflation by about 0.3% points 659 00:36:08,093 --> 00:36:10,925 per year from twenty nineteen to twenty twenty five. 660 00:36:11,425 --> 00:36:13,046 So the measurement's not just lagging, 661 00:36:13,046 --> 00:36:14,184 it's wrong by design. 662 00:36:14,684 --> 00:36:17,275 And if you're Megan trying to figure out why you can't afford your 663 00:36:17,275 --> 00:36:20,797 own childhood neighborhood, the official CPI number is telling 664 00:36:20,797 --> 00:36:22,728 you that housing inflation i is easing. 665 00:36:22,728 --> 00:36:25,109 But it isn't just wrong, it's insulting. 666 00:36:25,109 --> 00:36:28,971 It's telling you that your experience doesn't match the information, 667 00:36:28,971 --> 00:36:32,363 but the information isn't measuring your actual experience. 668 00:36:32,363 --> 00:36:35,585 It's measuring a phantom rent that literally no one pays. 669 00:36:36,085 --> 00:36:39,729 And when you're Chris Myers watching your wife's grocery dum budget double 670 00:36:39,729 --> 00:36:42,582 in a month because the government shut down and disrupted your paycheck, 671 00:36:42,582 --> 00:36:46,986 the CPI saying food inflation is two point seven percent isn't informative, 672 00:36:46,986 --> 00:36:47,986 it's noise. 673 00:36:48,673 --> 00:36:53,876 It is noise. And that noise is what's driving our current policy. 674 00:36:53,876 --> 00:36:56,718 The Federal Reserve is looking at their PCE number, 675 00:36:56,718 --> 00:36:58,399 the public's looking at the CPI, 676 00:36:58,399 --> 00:37:01,601 Chris and Megan are looking at their bills. 677 00:37:01,601 --> 00:37:04,913 None of those numbers are actually talking to one another and they're 678 00:37:04,913 --> 00:37:06,664 all pulled from different sources. 679 00:37:07,164 --> 00:37:12,277 And here's what that means. There's a pattern that goes back all the way to 1983, 680 00:37:12,277 --> 00:37:17,970 so long ago. The official measurement of inflation lags the lived cost, 681 00:37:17,970 --> 00:37:19,821 not by accident, by by design. 682 00:37:19,821 --> 00:37:23,633 When the BLS switched to owner's equivalent rent, 683 00:37:23,633 --> 00:37:27,816 the OER, they were making a deliberate choice to use a hypothetical number instead 684 00:37:27,816 --> 00:37:33,769 of an actual cost. When the the Boskin Commission adjusted that calculation in 1996, 685 00:37:33,769 --> 00:37:35,570 they were making a deliberate choice. 686 00:37:35,868 --> 00:37:39,740 To conclude that inflation was lower than previous calculations suggested, 687 00:37:39,740 --> 00:37:43,722 with the direct policy consequence of reducing Social Security. 688 00:37:43,722 --> 00:37:47,244 And then when the Fed switched to PCE in twenty twelve, 689 00:37:47,244 --> 00:37:50,916 they were making a deliberate choice to look at a different signal than 690 00:37:50,916 --> 00:37:56,188 the one the public hears. All three choices have technical justifications, 691 00:37:56,188 --> 00:37:59,140 but all three choices have consequences for real people. 692 00:37:59,640 --> 00:38:02,659 The consequence being that the policy is built on a number that doesn't match 693 00:38:02,659 --> 00:38:03,659 reality. 694 00:38:04,473 --> 00:38:07,534 Right. And here's why that's a problem beyond just, 695 00:38:07,534 --> 00:38:11,525 you know, the numbers wrong. If the Fed really believes that inflation 696 00:38:11,525 --> 00:38:15,016 is slowing down because they're looking at PCE, 697 00:38:15,016 --> 00:38:19,367 and inflation actually is running hotter than that because the real economy 698 00:38:19,367 --> 00:38:22,058 is more expensive than that can really capture, 699 00:38:22,058 --> 00:38:27,059 then they're gonna keep rates lower than the economic reality requires. 700 00:38:27,059 --> 00:38:30,590 They're gonna keep rates lower which 701 00:38:31,410 --> 00:38:34,618 makes borrowing cheaper, which makes more people demand things, 702 00:38:34,618 --> 00:38:37,355 which makes prices go higher, which makes inflation hotter, 703 00:38:37,355 --> 00:38:40,894 which is the opposite of what the policy goal actually is. 704 00:38:41,394 --> 00:38:43,127 So they're they're responding to the wrong signal. 705 00:38:43,627 --> 00:38:46,751 It might be. Or the signal's responding to them. 706 00:38:46,751 --> 00:38:50,676 If inflation stays high because the Fed's policy's off, 707 00:38:50,676 --> 00:38:53,299 then the Fed's going to point to that and say, 708 00:38:53,299 --> 00:38:56,743 see, inflation is sticky. And then they're going to rethink. 709 00:38:56,743 --> 00:38:59,986 But in the meantime, what's happening to the person actually paying for things? 710 00:39:00,486 --> 00:39:01,486 They're just paying. 711 00:39:01,763 --> 00:39:03,033 Their paycheck isn't going up, 712 00:39:03,033 --> 00:39:05,554 their rent, their mortgage, their groceries, 713 00:39:05,554 --> 00:39:08,195 all of those things are actually going up in real time. 714 00:39:08,195 --> 00:39:11,637 And when the official number says inflation is slowing down, 715 00:39:11,637 --> 00:39:14,808 and their lived experience says everything's getting more expensive, 716 00:39:14,808 --> 00:39:17,339 there's a gap between the official world and the real world. 717 00:39:17,339 --> 00:39:20,981 That gap makes us untrustful. 718 00:39:20,981 --> 00:39:25,432 It it it breaks trust. It it erodes the credibility of 719 00:39:25,432 --> 00:39:28,444 the institution that's supposed to be managing all of this. 720 00:39:28,444 --> 00:39:29,444 Our government. 721 00:39:29,988 --> 00:39:33,351 Or it's private bank the Fed. And more importantly, 722 00:39:33,351 --> 00:39:37,173 it means that the people making the decisions aren't actually connected 723 00:39:37,534 --> 00:39:40,286 to what those decisions cost the taxpayers. 724 00:39:40,786 --> 00:39:43,055 Which is important because the Fed has enormous power. 725 00:39:43,709 --> 00:39:48,161 Yes, they set the rates. Rates flow through the entire financial system, 726 00:39:48,161 --> 00:39:50,412 our entire economy is basically built on them. 727 00:39:50,412 --> 00:39:53,053 Every decision about money, borrowing, 728 00:39:53,053 --> 00:39:58,055 saving, investing, hiring, it all depends upon what the Fed decides. 729 00:39:58,055 --> 00:40:03,387 And they're deciding based on a metric that doesn't match the real lived inflation 730 00:40:03,387 --> 00:40:06,479 because of deliberate choices made in nineteen eighty three. 731 00:40:06,479 --> 00:40:11,171 Then they tweaked it in nineteen ninety six and then tweaked it again 732 00:40:11,171 --> 00:40:12,171 in twenty twelve. 733 00:40:12,485 --> 00:40:16,410 None of which anyone has asked the average taxpayer about, 734 00:40:16,410 --> 00:40:19,735 none of which were put to a democratic vote, 735 00:40:19,735 --> 00:40:25,423 and all of which are now embedded in how we measure the economy that when 736 00:40:25,423 --> 00:40:29,349 you challenge them it sounds like you're being weird instead of paying attention. 737 00:40:29,849 --> 00:40:31,961 Of course even inside the Fed there's doubt about whether 738 00:40:31,961 --> 00:40:33,610 the measurement's working accurately or not. 739 00:40:34,145 --> 00:40:37,586 There is. Kevin Walsh, who's the current Fed chair, 740 00:40:37,586 --> 00:40:41,867 actually testified to the Senate Banking Committee in July about exactly this thing. 741 00:40:41,867 --> 00:40:46,468 He said the Fed has established a task force to evaluate new data sources 742 00:40:46,468 --> 00:40:48,409 and consider methodological changes. 743 00:40:48,409 --> 00:40:53,550 And when he was asked, he asked them directly how how do we ensure that policymakers 744 00:40:53,550 --> 00:40:55,631 are receiving accurate, relevant, 745 00:40:55,631 --> 00:40:58,311 contemporaneous, actionable data? 746 00:40:58,311 --> 00:41:01,282 That's literally the Fed chair asking the Senate. 747 00:41:01,642 --> 00:41:04,175 Whether the data that he is using is any good. 748 00:41:04,175 --> 00:41:07,751 He's not confident. He's not saying the system is working. 749 00:41:07,751 --> 00:41:10,114 He's asking the Senate how to fix this. 750 00:41:10,614 --> 00:41:12,956 So we're at the point where even the Fed doesn't trust the data? 751 00:41:13,456 --> 00:41:17,188 Exactly. They're actually asking for a task force to figure out where there's better 752 00:41:17,188 --> 00:41:22,501 information. That means the entire money money policy apparatus is running 753 00:41:22,501 --> 00:41:26,573 on a measurement system that the people running it admit might 754 00:41:26,573 --> 00:41:30,025 not actually be working. And that's not trivial. 755 00:41:30,025 --> 00:41:33,376 That's the equivalent of an airline saying the the instruments 756 00:41:33,376 --> 00:41:35,237 on your airplane might not be accurate, 757 00:41:35,237 --> 00:41:36,248 but we're gonna fly anyway. 758 00:41:36,748 --> 00:41:41,305 Huh. ~ and the gap between this official number and the real number 759 00:41:41,305 --> 00:41:45,422 is only giving you feel more obvious with the higher inflation stays, 760 00:41:45,422 --> 00:41:46,422 right? 761 00:41:46,219 --> 00:41:47,219 Yeah. 762 00:41:46,699 --> 00:41:48,700 Right now people are frustrated. 763 00:41:48,700 --> 00:41:52,102 I I would say more than that. They they feel that gap every day. 764 00:41:52,102 --> 00:41:55,684 But if inflation stays high or it gets worse, 765 00:41:55,684 --> 00:41:57,905 then that gap is gonna be undeniable. 766 00:41:57,905 --> 00:42:02,477 The Fed's gonna be pointing their PCE numbers and saying inflation's under control 767 00:42:02,477 --> 00:42:05,449 and people are gonna be paying their bills and knowing that that's just not true. 768 00:42:05,449 --> 00:42:08,511 And the institutions are gonna lose credibility, 769 00:42:08,511 --> 00:42:11,253 not because they're corrupt, mm they are, 770 00:42:11,253 --> 00:42:14,975 but because they're literally just not measuring the same world that people 771 00:42:14,975 --> 00:42:15,975 are living in. 772 00:42:15,965 --> 00:42:19,778 So here's what's I guess the wildest part of this whole story. 773 00:42:19,778 --> 00:42:22,560 It's the thing that holds the whole thing up. 774 00:42:22,560 --> 00:42:27,635 It's that the government and the government's private central bank 775 00:42:27,635 --> 00:42:31,508 are measuring inflation with formulas instead of experience. 776 00:42:31,508 --> 00:42:33,990 They're not measuring what a house costs, 777 00:42:33,990 --> 00:42:36,612 but what you'd charge yourself to rent your own house, 778 00:42:36,612 --> 00:42:39,775 not measuring what people actually bought yesterday, 779 00:42:39,775 --> 00:42:43,458 but what would they buy if prices didn't change? 780 00:42:43,958 --> 00:42:45,536 It's mathematically defensible, 781 00:42:45,536 --> 00:42:48,692 but completely disconnected from how any of us actually budget. 782 00:42:49,192 --> 00:42:52,100 And the person making decisions off of these numbers is either looking 783 00:42:52,100 --> 00:42:55,289 at a different number still, or they're looking at the same number, 784 00:42:55,289 --> 00:42:57,876 but it's running six months behind what's actually happening in the world. 785 00:42:58,650 --> 00:43:00,962 Right, so pay attention to this. 786 00:43:00,962 --> 00:43:04,936 Pay attention to the gap. Don't just look at the CPI headline, 787 00:43:04,936 --> 00:43:07,528 look at the cost of housing, look how fast it's moving. 788 00:43:07,528 --> 00:43:09,761 Look at what the Fed is actually targeting, 789 00:43:09,761 --> 00:43:13,374 look at the PCE numbers because those two numbers are moving apart right now. 790 00:43:13,374 --> 00:43:18,879 They're moving apart because one of them is real and one of them is theoretical. 791 00:43:18,879 --> 00:43:22,823 And the policy's built on the theoretical. 792 00:43:23,323 --> 00:43:26,867 So one concrete thing people can actually do next time there's a CPI report 793 00:43:27,248 --> 00:43:28,769 is to look at the shelter component. 794 00:43:28,769 --> 00:43:30,461 And not just read the headline, 795 00:43:30,461 --> 00:43:31,915 listen to what the news is saying about it, 796 00:43:31,915 --> 00:43:34,759 but actually look at whether it's accelerating or decelerating. 797 00:43:34,759 --> 00:43:37,592 Because the shelter keeps running hot and the official story 798 00:43:37,592 --> 00:43:41,046 is that inflation's easing, you're seeing the lie in real time. 799 00:43:41,546 --> 00:43:45,608 And you should ask yourself why we're measuring the thing you pay for with 800 00:43:45,608 --> 00:43:47,188 a number and nobody actually pays. 801 00:43:47,188 --> 00:43:52,370 That one question reveals the entire logic of why the system works the way it does. 802 00:43:52,370 --> 00:43:55,631 We measure inflation with formulas because formulas are stable. 803 00:43:55,631 --> 00:44:00,593 Formulas don't change with politics or what president is in the Oval Office. 804 00:44:00,593 --> 00:44:05,495 And it sounds like science. If the formula doesn't match your life, 805 00:44:05,495 --> 00:44:07,776 well maybe your life is wrong. 806 00:44:07,776 --> 00:44:09,727 But that's not how it works. That's not 807 00:44:10,273 --> 00:44:12,224 How they want you to think it works at least. 808 00:44:12,724 --> 00:44:16,427 So listeners as you know the Overlap is a show about systems of power, 809 00:44:16,427 --> 00:44:18,388 labor, and American injustice. 810 00:44:18,388 --> 00:44:21,751 You can find us at the FOF.foundation, 811 00:44:21,751 --> 00:44:24,653 and you can follow us on Blue Sky or Mastodon. 812 00:44:25,153 --> 00:44:28,497 Yeah, also don't forget to share interesting posts with your friends and family. 813 00:44:28,497 --> 00:44:30,419 Shoot us an email at the overlap. 814 00:44:30,419 --> 00:44:33,102 The overlap at FOF.foundation. 815 00:44:33,102 --> 00:44:36,385 And make sure you give us a five star rating on the podcast 816 00:44:36,385 --> 00:44:38,047 app you're currently listening to, 817 00:44:38,047 --> 00:44:40,750 like Apple, Spotify, Podveen, whatever. 818 00:44:40,750 --> 00:44:43,293 It helps us reach more engaged listeners like you. 819 00:44:43,793 --> 00:44:46,193 The more you know. Bye. 820 00:44:44,010 --> 00:44:47,360 And with that, bye.