WEBVTT - How a Secretive Conclave Decides When U.S. Recessions Happen 0:00:00.240 --> 0:00:04.640 Last month, December twenty seventeen, marked the ten year anniversary 0:00:04.640 --> 0:00:08.280 of an economic event that probably won't be celebrated anywhere. 0:00:08.600 --> 0:00:11.200 It was the beginning of the deepest recession in the 0:00:11.320 --> 0:00:15.480 United States since the Great Depression in the nineteen thirties. Now, 0:00:15.640 --> 0:00:17.960 we know the economy was in pretty bad shape at 0:00:18.000 --> 0:00:20.800 that time, thanks to the bursting of the housing bubble 0:00:21.120 --> 0:00:24.479 and the early stages of the global financial crisis. But 0:00:24.600 --> 0:00:28.000 how do we know that the recession began precisely in 0:00:28.080 --> 0:00:31.720 December two thousand seven. Today we're going to talk about 0:00:31.760 --> 0:00:36.000 the mysterious process that determines when recessions begin and when 0:00:36.040 --> 0:00:49.199 they end. Welcome to our first Benchmark episode of eighteen. 0:00:49.600 --> 0:00:53.320 I'm Scott landman and economics editor with Bloomberg News in Washington. 0:00:53.800 --> 0:00:57.280 Now I'm Daniel Moss, economics writer and editor at Bloomberg 0:00:57.360 --> 0:01:01.720 View in New York. So, and it's pretty commonly thought 0:01:01.800 --> 0:01:05.319 that two quarters of declining g DP means that a 0:01:05.360 --> 0:01:08.280 recession has begun. But that's not the case in the 0:01:08.360 --> 0:01:12.400 United States, right, That's right. The US is really an outlier, 0:01:12.720 --> 0:01:16.760 certainly in terms of major economies. What most people might 0:01:16.840 --> 0:01:20.319 not know is when it comes to deciding when a 0:01:20.400 --> 0:01:23.959 slump starts and ends, it's up to a small group 0:01:24.080 --> 0:01:27.880 of elite professors from around the country who deliberate in 0:01:27.959 --> 0:01:31.320 secret and then announced their decision to the world. It's 0:01:31.319 --> 0:01:34.160 a little like waiting for the white smoke from the 0:01:34.240 --> 0:01:38.080 Vatican chimney. It's also a lot like those monetary policy 0:01:38.120 --> 0:01:41.080 meetings we cover all the time at Bloomberg, except this 0:01:41.120 --> 0:01:44.679 group makes decisions about events that happened months even a 0:01:44.760 --> 0:01:48.720 year before. In fact, when the recession began in December 0:01:48.720 --> 0:01:51.960 two thousand seven, it wasn't until a full year later 0:01:52.000 --> 0:01:55.280 that this committee announced that that was the actual date 0:01:55.360 --> 0:02:00.440 when the recession officially began. So who are these people? Well, 0:02:00.520 --> 0:02:03.080 our guest today can tell us all about that. His 0:02:03.200 --> 0:02:06.080 name is Robert Hall, and he's a professor of economics 0:02:06.120 --> 0:02:10.000 at Stanford University in California. He is also the chair 0:02:10.040 --> 0:02:13.680 of the Business Cycle Dating Committee of the National Bureau 0:02:13.800 --> 0:02:16.920 of Economic Research, and that's the group that tells the 0:02:16.919 --> 0:02:20.480 world when a US recession begins and ends. He's joining 0:02:20.560 --> 0:02:23.240 us by phone here in Washington, where he's en route 0:02:23.280 --> 0:02:27.680 to an annual meeting of economists in Philadelphia this weekend. Bob, 0:02:27.840 --> 0:02:31.960 thanks for joining us on benchmark. So, Bob, can you 0:02:32.000 --> 0:02:33.920 give us a little history. First of all, how is 0:02:33.960 --> 0:02:37.720 this committee formed and how did it become the accepted 0:02:37.840 --> 0:02:42.840 arbiter of when recessions begin and end in the United States? Sure, well, 0:02:42.880 --> 0:02:44.880 we have to go right back. The National Bureau was 0:02:44.919 --> 0:02:50.840 created in starting the beginning of a great depression. It 0:02:50.960 --> 0:02:57.040 took on this job itself just spontaneously to decide when 0:02:57.480 --> 0:03:02.400 recessions began and ended. It has done someone informally until 0:03:04.320 --> 0:03:08.600 when Martin Feldstein took over the National Bureau and appointed 0:03:08.639 --> 0:03:11.600 me to chair this committee and kind of formalized the 0:03:11.600 --> 0:03:15.959 whole process. So, for example, we know, write an explanatory memo, 0:03:16.440 --> 0:03:18.680 and that memo has made public at a certain time, 0:03:19.320 --> 0:03:21.520 usually about a year after or maybe a year and 0:03:21.520 --> 0:03:25.000 a half after the actual event. So and we've been 0:03:25.000 --> 0:03:27.960 doing that consistently pretty much the same basis ever since. 0:03:28.480 --> 0:03:34.000 And the name National Bureau of Economic Research sounds federal 0:03:34.080 --> 0:03:37.720 in Nightsha. Does it have any relationship with the government. No, 0:03:37.920 --> 0:03:42.080 not a formal relationship. Many people, for example, Feldstine himself 0:03:42.160 --> 0:03:45.440 was Chamber of the Council of Economic Advisors. But the 0:03:45.520 --> 0:03:50.640 National Bureau is a the inter university organization as a 0:03:51.680 --> 0:03:56.680 governing body that universities contribute directors too, and it sponsors research, 0:03:56.720 --> 0:04:01.440 and it sponsors many many conferences and publications. But it's 0:04:01.440 --> 0:04:05.960 basically an academic organization based at universities. Now, how long 0:04:06.000 --> 0:04:09.880 have you been on the Business Cycle Dating Committee and 0:04:09.960 --> 0:04:14.200 been its chairman as well? What's wrong with the way 0:04:14.360 --> 0:04:18.440 many countries do this, Bob? Where I'm from Australia and 0:04:18.480 --> 0:04:21.719 it's by no means limited to Australia. It's easy to 0:04:21.800 --> 0:04:24.919 tell when a recession arrives and ends because there's a 0:04:24.960 --> 0:04:28.560 standard definition of two quarters of contraction in g d 0:04:28.720 --> 0:04:34.280 P marks a recession. What's wrong with that definition? You know, 0:04:34.400 --> 0:04:35.920 a lot of the time, there's nothing wrong with that. 0:04:36.120 --> 0:04:40.000 If you look at our chronology, it frequently coincides with that. 0:04:40.640 --> 0:04:43.240 There's some common sense to that idea. But we are 0:04:43.320 --> 0:04:47.000 sensitive to certain possibilities have arisen. We've had to make 0:04:47.120 --> 0:04:51.440 judgment calls. For example, if there's a pause in the 0:04:51.480 --> 0:04:56.480 economy but it doesn't have the drama and depth of 0:04:56.520 --> 0:04:58.880 what we would call a recession, but it lasts at 0:04:58.880 --> 0:05:01.160 two quarters. So if we had it just a couple 0:05:01.160 --> 0:05:05.040 of quarters where the economy contracted by less than a 0:05:05.080 --> 0:05:08.560 percentage point, each probably wouldn't call that the recession. We 0:05:08.600 --> 0:05:11.440 would just just included in whatever else was going on 0:05:11.480 --> 0:05:14.920 at the time, either a contraction or expansion. So we're 0:05:14.920 --> 0:05:18.919 alert to that possibility. To be honest, it actually hasn't 0:05:18.920 --> 0:05:22.400 arisen very dramatically, and you wouldn't go too far off 0:05:23.200 --> 0:05:26.200 if you adopted that to quarter. We're not we're not 0:05:26.279 --> 0:05:29.640 totally against it and saying it's totally false, but we 0:05:29.720 --> 0:05:33.000 approached this was a little more nuanced, but you probably have. 0:05:33.120 --> 0:05:36.279 You have more precise data in the United States, probably 0:05:36.279 --> 0:05:38.559 a richer data set than most countries in the world. 0:05:38.839 --> 0:05:42.080 We have jobs data in this country that are available 0:05:42.080 --> 0:05:44.640 on a on a very good level of detail. Does 0:05:44.640 --> 0:05:49.679 that help you pinpoint the month where business cycle peaks? Yes. Actually, 0:05:49.720 --> 0:05:52.320 you mentioned one feature that is important, and that is 0:05:52.800 --> 0:05:56.200 we do want to identify turning points in a particular month, 0:05:56.920 --> 0:06:00.240 and the totality of GDP is available only on a 0:06:00.320 --> 0:06:03.960 quarterly basis, so we'd like to fine tune our work, 0:06:04.360 --> 0:06:07.120 and we do that using many different monthly series that 0:06:07.200 --> 0:06:11.479 you mentioned employment, many of the components is output are 0:06:11.480 --> 0:06:15.680 now measured on a monthly basis, even though the GDP 0:06:15.839 --> 0:06:18.760 number itself is quarterly so we look at that too. 0:06:19.320 --> 0:06:22.680 I guess this also gets to some of the challenges 0:06:22.760 --> 0:06:26.080 that people have identified when it comes to measuring economic 0:06:26.120 --> 0:06:29.800 well being purely on the basis of GDP. What else 0:06:29.800 --> 0:06:32.120 do you look at, Bob Well, First of all, let's 0:06:32.160 --> 0:06:36.200 be very clear that we don't claim to be measuring 0:06:36.800 --> 0:06:40.560 the movements of social welfare, and that's an interesting, big topic. 0:06:41.360 --> 0:06:45.200 But we deal with something we call economic activity, which 0:06:45.240 --> 0:06:49.200 in particularly we mean market activity. We're focusing on the 0:06:49.240 --> 0:06:52.800 amount of output that the economy produces and the number 0:06:52.839 --> 0:06:56.520 of jobs that it delivers, but we don't try to 0:06:56.560 --> 0:06:58.840 go beyond that and say that we're doing more than that. 0:06:59.160 --> 0:07:01.920 So I think we should be very clear that those 0:07:02.240 --> 0:07:06.200 more technical measures that we're dealing with. Of course, output 0:07:06.200 --> 0:07:09.680 and jobs are very important. So people are very interested 0:07:09.800 --> 0:07:13.240 by the way in the outcome of our deliberations. We're 0:07:13.280 --> 0:07:15.640 doing something that they're interested in, even though we don't 0:07:15.680 --> 0:07:19.400 say that it measures the totality of welfare. So let's 0:07:19.440 --> 0:07:22.680 go back in time ten years ago. It's December two 0:07:22.720 --> 0:07:27.200 thousand seven. We already know the economy is in fairly 0:07:27.280 --> 0:07:31.080 bad shape at that time, could be in recession already. 0:07:32.120 --> 0:07:35.480 But your committee isn't going to announce a decision until 0:07:35.640 --> 0:07:39.640 about a year later on when the recession actually began. 0:07:40.320 --> 0:07:44.880 What is your committee doing in that time to figure 0:07:44.880 --> 0:07:47.160 it out? What kind of deliberations do you have? Can 0:07:47.200 --> 0:07:49.480 you walk us through that at all? Well, the committee 0:07:49.840 --> 0:07:53.760 the modern era has been based primarily on email. The 0:07:53.800 --> 0:07:58.440 email attachment is the way we deliberate more than anything else. 0:07:59.160 --> 0:08:02.400 And when it looks like the economies in some trouble, 0:08:02.640 --> 0:08:07.080 as it did in late two thousand seven, then informally, 0:08:07.520 --> 0:08:11.280 without necessarily convening something you should call a meeting, we 0:08:11.280 --> 0:08:17.200 start trading data almost entirely through email. Then as things 0:08:17.320 --> 0:08:21.280 come into sharper focus, we generally convene a meeting by 0:08:21.320 --> 0:08:24.320 a conference call. And at that point we've all seen 0:08:24.400 --> 0:08:28.880 each other's ideas about what's happening, and then we deliberate, 0:08:29.400 --> 0:08:31.600 and then finally, usually a year to a year and 0:08:31.600 --> 0:08:34.520 a half later, we have a meeting which actually decides 0:08:34.720 --> 0:08:39.000 when the turning point occurred, the peak of activity, which, 0:08:39.000 --> 0:08:41.839 as you say, we determined occurred in December of two 0:08:41.880 --> 0:08:45.880 thousand seven, we write up a memo. The memo is 0:08:45.920 --> 0:08:49.400 posted on the National Bureau website. It's released to it 0:08:49.480 --> 0:08:53.520 Financial Press and usually as front page news. Well, are 0:08:53.520 --> 0:08:57.880 you ever worried about the security of these email communications? 0:08:58.160 --> 0:09:00.559 It's certainly something we thought about, is in the more 0:09:00.600 --> 0:09:06.640 general context of making sure that we don't create opportunities 0:09:06.920 --> 0:09:10.840 for people to break into the process. We use encrypted email, 0:09:11.320 --> 0:09:16.920 and we use completely secure telephone calls, and the National 0:09:16.960 --> 0:09:20.240 Bureau is a very good record of maintaining the records 0:09:20.240 --> 0:09:23.120 of this process, which are done at National Bureau headquarters. 0:09:23.880 --> 0:09:26.960 So it's a theoretical possibility. It certainly is not one 0:09:27.000 --> 0:09:29.600 that's that's created any problem so far, and it's something 0:09:29.640 --> 0:09:33.480 regarding against the future. And none of you, of course 0:09:33.920 --> 0:09:36.800 strangers to each other, all to the way you see 0:09:36.840 --> 0:09:42.440 the world. You've mentioned the American Economics Association meeting in Philadelphia. 0:09:42.600 --> 0:09:45.280 You do all see each other around the traps, if 0:09:45.360 --> 0:09:47.439 not there at the I m F World Bank and 0:09:47.480 --> 0:09:50.959 other places. Right, sure, not to mention National Bureau conferences, 0:09:50.960 --> 0:09:54.400 which are a very big part of the deal. Let's 0:09:54.440 --> 0:09:56.560 talk a little bit more about the membership of the committee. 0:09:56.559 --> 0:09:58.920 I mean, I see some of the names on your committee, 0:09:58.960 --> 0:10:01.400 like Martin Felts you've mentioned it. It looks to me 0:10:01.480 --> 0:10:04.120 like an economic all star team. I mean, maybe the 0:10:04.200 --> 0:10:07.160 names aren't widely known to the general public, but many 0:10:07.200 --> 0:10:09.720 of them are our household names in the world of 0:10:10.080 --> 0:10:15.560 academia economics departments. How do you go about picking the committee? 0:10:15.640 --> 0:10:18.480 Is there any way I can be on it? I 0:10:18.920 --> 0:10:21.600 know you have to have a track record as an 0:10:21.640 --> 0:10:27.040 academic macro economist, as you have to specialize in the 0:10:27.160 --> 0:10:31.160 study of the subject at hand, which is what happens 0:10:31.240 --> 0:10:34.640 to have put unemployment in other other economy wide variables. 0:10:34.840 --> 0:10:37.200 So we're looking very much at the economy wide movements, 0:10:37.760 --> 0:10:42.480 and the subject of macroeconomics basically covers that. So everyone 0:10:42.520 --> 0:10:46.720 you see here is recognized as a specialist in macroeconomics. 0:10:47.000 --> 0:10:49.000 Are people beating down your door to get on the 0:10:49.000 --> 0:10:51.320 committee or is it kind of accepted practice that you 0:10:51.440 --> 0:10:54.240 come to them? Uh? No, we come to them exactly. Yeah. 0:10:54.840 --> 0:10:58.800 It would be very unfortunate if somebody called me up, 0:10:58.920 --> 0:11:01.720 or called up James Pitterbin said I want to be 0:11:01.720 --> 0:11:03.679 on the committee. That would be That's just not the 0:11:03.720 --> 0:11:06.760 way people behave They wait for us to call them. 0:11:06.840 --> 0:11:09.720 So it is like a papal conclave. You know, I've 0:11:09.720 --> 0:11:14.640 never thought about that, but yes, in that way, it's 0:11:14.679 --> 0:11:17.720 not it's not quite as dramatic. You know, we don't 0:11:17.720 --> 0:11:22.000 meet physically. No one would consider electing the pope on 0:11:22.040 --> 0:11:24.400 the basis conference call. But that's the way we work, 0:11:24.720 --> 0:11:27.360 an email conclave. That that sounds like something for the 0:11:27.400 --> 0:11:33.040 modern era exactly, you know, going by the NBAS definition, 0:11:33.520 --> 0:11:36.840 this current expansion, that worrying is starting to get pretty 0:11:36.840 --> 0:11:40.160 long in the tooth historically. Yes, yes, you're absolutely right. 0:11:40.679 --> 0:11:44.040 And without going into any of the committee's deliberations, which 0:11:44.240 --> 0:11:48.120 we know you can't, are you seeing anything right now 0:11:48.280 --> 0:11:52.040 in our current economic trajectory the United States that gives 0:11:52.080 --> 0:11:55.840 you pause? Oh well, it's easy to think of things 0:11:55.880 --> 0:11:58.319 that give paused, but you know, it's very important to 0:11:58.400 --> 0:12:01.319 understand that we don't operate on a forward looking basis. 0:12:01.840 --> 0:12:04.240 So I guess one thing I can say is that 0:12:04.320 --> 0:12:06.760 there's been no desire in the part of the committee 0:12:06.760 --> 0:12:10.600 to start the process that I described. It's the committee 0:12:10.640 --> 0:12:14.400 right now is everyone's watching. But the macroeconomic news has 0:12:14.440 --> 0:12:17.160 been pretty good lately. And but you know, I would 0:12:17.160 --> 0:12:18.640 have said the same thing at the beginning of two 0:12:18.640 --> 0:12:21.840 thousand seven. So it's been surprising to me in my 0:12:22.040 --> 0:12:25.720 role to see how quickly things change and they begin 0:12:25.800 --> 0:12:28.079 to be concerns and then and then this process that 0:12:28.160 --> 0:12:31.280 I described is set in motion. Is that because the 0:12:31.360 --> 0:12:34.760 link between capital markets and what might be called the 0:12:34.880 --> 0:12:40.280 real economy is ever closer and closer and closer. No, 0:12:40.360 --> 0:12:42.560 I wouldn't say that, because you know, I've been doing 0:12:42.559 --> 0:12:45.840 this job for a long time to quite a few cycles, 0:12:46.320 --> 0:12:49.000 and I think the timing has remained pretty much the same. 0:12:49.440 --> 0:12:53.000 The clouds can gather quite quickly, and I don't think 0:12:53.040 --> 0:12:57.040 that's new. We didn't have a severe the previous severe 0:12:57.160 --> 0:12:59.640 session before the one that began to do that, and 0:12:59.679 --> 0:13:05.080 seven was in two it was almost as severe, and 0:13:05.160 --> 0:13:09.559 it came on quite surprisingly quickly. So but you're right 0:13:09.600 --> 0:13:12.679 that there is a close connection between financial events and 0:13:12.800 --> 0:13:16.360 the onset of a recession. There's there's no doubt, I think, 0:13:16.400 --> 0:13:19.000 in terms of the most analysts that the procession that 0:13:19.040 --> 0:13:20.920 began at the end of two thousand seven was made 0:13:21.000 --> 0:13:25.240 much worse the financial crisis that occurred in September two eight. 0:13:26.040 --> 0:13:29.200 Speaking of financial events, or maybe fiscal events, we can't 0:13:29.240 --> 0:13:32.760 let you go without asking you a question about a 0:13:32.800 --> 0:13:36.720 major economics story. Just a couple of weeks ago, a 0:13:36.760 --> 0:13:41.720 significant piece of tax legislation was signed into law economists 0:13:41.760 --> 0:13:44.320 in general have said it might give a modest boost 0:13:44.360 --> 0:13:48.079 to the economy in maybe a little bit in twenty nine, 0:13:48.480 --> 0:13:53.000 but most people also agree that it'll significantly widened the 0:13:53.080 --> 0:13:57.000 budget deficit and potentially raise economic risks. What's your take 0:13:57.040 --> 0:14:00.360 on how this law will affect the economy and doesn't 0:14:00.440 --> 0:14:03.000 raise the risk of a recession happening and say a 0:14:03.000 --> 0:14:05.680 couple of years from now. Oh hard to say. I 0:14:05.720 --> 0:14:09.520 think in my capacity is chairman of this committee, that's 0:14:09.559 --> 0:14:12.360 not the type of thing we consider. We wait for 0:14:12.400 --> 0:14:16.640 things to happen. We're definitely not in the forecasting business. Obviously, 0:14:16.880 --> 0:14:20.840 I have another life as somebody who thinks about topics 0:14:20.880 --> 0:14:24.920 like that. But as far as my representing this committee, 0:14:25.080 --> 0:14:26.400 the thing for me to say is we're not in 0:14:26.400 --> 0:14:30.000 the forecasting business or very much in the what happened 0:14:30.000 --> 0:14:32.440 over the past previous year, not what's going to happen 0:14:32.480 --> 0:14:35.000 next year. That's that's my job on the top of 0:14:35.000 --> 0:14:38.040 the committee. Bob, thanks very much for being with us 0:14:38.040 --> 0:14:44.880 today on Benchmark my pleasure. Benchmark will be back next week. 0:14:45.000 --> 0:14:47.320 Until then, you can find us on the Bloomberg terminal, 0:14:47.320 --> 0:14:50.160 Bloomberg dot com or Bloomberg app as well as wherever 0:14:50.200 --> 0:14:55.040 you enjoy podcasts, including Apple Podcasts, overcasting Stitcher. While you're there, 0:14:55.160 --> 0:14:57.000 take a minute to rate and review the show so 0:14:57.080 --> 0:14:59.280 more listeners can find us and let us know what 0:14:59.320 --> 0:15:01.440 you thought of the show. You can follow me on 0:15:01.480 --> 0:15:06.800 Twitter at at scott Landman Dan you're at Moss Underscore. Ka. 0:15:07.400 --> 0:15:10.040 Bob isn't on Twitter, but you can follow releases from 0:15:10.040 --> 0:15:12.640 the n B E r at at n B E 0:15:13.000 --> 0:15:16.760 r R e L. Benchmark is produced by Topher Foreheads 0:15:17.080 --> 0:15:20.480 and the head of Bloomberg Podcasts is Francesca Levy. Thanks 0:15:20.480 --> 0:15:21.840 for listening, See you next time.