WEBVTT NOTE The Best Arguments for Trade Protection — and Why They Are Wrong 1 00:00:00.000 --> 00:00:05.800 My paper is about the best arguments for protection, and by protection I'm talking about 2 00:00:05.800 --> 00:00:12.240 protecting ourselves, protecting domestic industries from foreign competition with some type of trade restrictions. 3 00:00:12.240 --> 00:00:22.800 And there's an awful lot of arguments in favor of protection, and I don't want to have time to talk about all of them, but in the next few minutes I'd like to 4 00:00:22.800 --> 00:00:34.060 say that a useful way to think about these arguments is to put them into two categories. There are some naive arguments in favor of protection and then there are the best arguments in favor of protection. 5 00:00:34.060 --> 00:00:43.960 And then I'll have some comments about whether or not we should accept the best arguments. 6 00:00:43.960 --> 00:00:49.560 The case for free trade starts with the law of comparative advantage, but it doesn't end there. 7 00:00:49.560 --> 00:00:51.560 And I assume we're familiar with this law. 8 00:00:51.560 --> 00:01:07.560 Ricardo's law of comparative advantage says that if we specialize in producing the products where we have a comparative advantage and export those products and import the products where the other country has a comparative advantage, then this will increase social product. 9 00:01:07.560 --> 00:01:12.560 And this is an airtight argument, but it only holds under certain conditions. 10 00:01:12.560 --> 00:01:15.560 Within the conditions, it's an airtight argument. 11 00:01:15.560 --> 00:01:23.560 So the explicit premise of that argument is that capital is immobile across political borders. 12 00:01:23.560 --> 00:01:30.560 It's mobile within the political borders, but capital and other resources do not flow between countries. 13 00:01:30.560 --> 00:01:38.560 There are other sort of, they're not really premises to the law of comparative advantage, but they're sort of conditions. 14 00:01:38.560 --> 00:01:56.560 And Ricardo and everyone who has espoused this argument in the last two centuries recognize that they're not talking about situations where there's externalities or monopoly power, and that the law doesn't necessarily hold in those cases, and maybe in some other cases. 15 00:01:56.560 --> 00:02:07.560 So the naive arguments, which by the way are oftentimes the powerful arguments and they tend to hold the most weight in public opinion, 16 00:02:07.560 --> 00:02:11.560 the naive arguments are the ones that contradict the law of comparative advantage. 17 00:02:11.560 --> 00:02:17.560 So there are arguments out there that say we should protect ourselves from inexpensive imports, 18 00:02:17.560 --> 00:02:23.560 we should protect ourselves from trade with countries that have cheap foreign labor, 19 00:02:23.560 --> 00:02:27.560 We should protect ourselves from labor or we should protect ourselves from dumping. 20 00:02:27.560 --> 00:02:38.560 We should protect ourselves from allowing foreign firms selling their products, importing their products to us at prices that are relatively low in order to gain market share. 21 00:02:38.560 --> 00:02:46.560 There are also arguments about we should protect ourselves in order to increase jobs, increase production or increase employment. 22 00:02:46.560 --> 00:02:52.560 And there are arguments that say we should protect ourselves in order to foster infant industries. 23 00:02:52.560 --> 00:03:01.560 And all of these arguments by themselves are wrong because they contradict the law of comparative advantage. 24 00:03:01.560 --> 00:03:09.560 The proponents of these arguments see the benefits of the protection, but they don't see what Bastiat says. 25 00:03:09.560 --> 00:03:11.560 They don't see what is unseen. 26 00:03:11.560 --> 00:03:19.560 And we can dispose of these arguments relatively quickly, but there are popular arguments out there. 27 00:03:19.560 --> 00:03:26.560 Let me say one thing. Let me add one thing to that, though. A lot of the infant industries arguments, 28 00:03:26.560 --> 00:03:32.560 the infant industries arguments is that we need to protect industries so that they will mature and become competitive industries 29 00:03:32.560 --> 00:03:35.560 and the type of industries we want to have in the long run. 30 00:03:35.560 --> 00:03:40.560 Some of those infant industries arguments are mixed in with more complicated arguments 31 00:03:40.560 --> 00:03:52.560 that include conditions that don't hold under Ricardo's law of comparative advantage. 32 00:03:52.560 --> 00:03:55.560 So there are more sophisticated infant industries arguments out there. 33 00:03:55.560 --> 00:04:01.560 I don't have a lot of time to dispose of these naive arguments. 34 00:04:01.560 --> 00:04:09.560 I trust that you see that the proponents of these arguments are simply not seeing the whole picture here. 35 00:04:09.560 --> 00:04:21.560 Okay, what I'm calling the best arguments for protection are the arguments where the conditions under which the law of comparative advantage hold are no longer the case. 36 00:04:21.560 --> 00:04:31.600 So, there's a series of arguments out there in the trade literature about protecting ourselves 37 00:04:31.600 --> 00:04:37.720 or protecting industries that generate positive externalities, or maybe diminishing industries 38 00:04:37.720 --> 00:04:41.780 that have negative externalities. But generally they're talking about maybe we should protect 39 00:04:41.780 --> 00:04:46.540 industries that would generate positive externalities for the domestic country. And most of this 40 00:04:46.540 --> 00:04:51.940 This literature focuses on research and development. So the idea is if an industry is going to 41 00:04:51.940 --> 00:04:56.300 be good at research and development, we want it here, we should protect ourselves so that 42 00:04:56.300 --> 00:05:00.220 that industry generates positive externalities for the rest of the country. So even though 43 00:05:00.220 --> 00:05:04.820 that particular industry doesn't capture the gains, other industries in the country will 44 00:05:04.820 --> 00:05:11.820 capture those gains. One problem with that, I was going to go through this whole list, 45 00:05:11.820 --> 00:05:15.520 I was going to go through this whole list, but one problem with that is that these types 46 00:05:15.520 --> 00:05:18.200 of externalities tend to cross political borders. 47 00:05:18.200 --> 00:05:23.740 So in the neoclassical modeling of these arguments, they tend to say positive externalities stop 48 00:05:23.740 --> 00:05:26.100 at the political borders. 49 00:05:26.100 --> 00:05:31.860 But in today's world, you would think that if a firm in Asia developed a new technology, 50 00:05:31.860 --> 00:05:37.020 that firms in the US would be quick to embrace those new technologies. 51 00:05:37.020 --> 00:05:39.940 Regarding research and development, let me mention Gottfried Habler's. 52 00:05:39.940 --> 00:05:47.940 If you were going to read a book on international trade theory, I recommend Gottfried Hobbler's 53 00:05:47.940 --> 00:05:54.220 Theory of International Trade. It's a 1936 book, at least the first English edition is 54 00:05:54.220 --> 00:06:02.620 in 1936. But Hobbler argues that in cases of research and development, he talks about 55 00:06:02.620 --> 00:06:13.020 and Educative Effect. He argues that we should allow the country that has a comparative advantage 56 00:06:13.020 --> 00:06:18.980 in research and development to do that research and development and develop the new technologies 57 00:06:18.980 --> 00:06:22.940 and then we should have free trade and then free trade will generate a lot of competition 58 00:06:22.940 --> 00:06:28.840 here and that competition will force us to embrace the new technologies. So Hobbler turns 59 00:06:28.840 --> 00:06:38.840 In the case of research and development, the argument should be that we should adopt free trade instead of protection. 60 00:06:38.840 --> 00:06:50.840 Possibly better arguments for protection are the arguments where there's imperfect competition in markets, where there's monopoly rents. 61 00:06:50.840 --> 00:07:08.840 And so there's all of these models out there where they say it's a second best world, there are monopolies present, and then in a second best world you can develop a model that draws about any conclusion you want. 62 00:07:08.840 --> 00:07:18.840 So, for instance, some of the conclusions are that the optimal policy is what they call a negative tariff, so that you should subsidize other countries to send you stuff. 63 00:07:18.840 --> 00:07:26.840 You should take tax dollars in the face of imperfect competition and give it to other countries so that they'll send you products, which doesn't... 64 00:07:26.840 --> 00:07:30.840 But under the right modeling conditions, that's the right option. 65 00:07:30.840 --> 00:07:40.840 But regarding monopoly rents, many of the papers fall into two categories. Many of the arguments fall into two categories. 66 00:07:40.840 --> 00:07:48.840 And one is, if there's going to be a firm that generates monopoly rents, we should protect that firm so that it will locate in this country. 67 00:07:48.840 --> 00:07:56.840 And then the firm will sell products overseas, foreign consumers will pay the monopoly rents, and this country will capture the monopoly rents. 68 00:07:56.840 --> 00:08:09.840 The reverse argument is that if the foreign monopolist is going to be overseas and going to export into this country, we should enact a tariff so that our government captures the monopoly rents. 69 00:08:09.840 --> 00:08:19.840 That foreign monopolist pays the rents in the form of tariff revenue to our government. So we extract, they call it extracting the rents from the foreign monopolist. 70 00:08:19.840 --> 00:08:29.760 Let me mention one more argument and then I'll say why I think all of these best arguments 71 00:08:29.760 --> 00:08:36.360 are why we should be suspicious of these arguments. 72 00:08:36.360 --> 00:08:43.520 The third best argument, it's not the third best, the third argument in this category 73 00:08:43.520 --> 00:08:49.800 is the argument where capital is flowing across political borders, capital is mobile. 74 00:08:49.800 --> 00:08:57.600 And I think this is the main argument today. I think this is the argument we hear if you 75 00:08:57.600 --> 00:09:02.960 listen closely. I think this is the argument, a few years ago at the Austrian Scholars Conference, 76 00:09:02.960 --> 00:09:07.760 Paul Crete Roberts gave a talk, and I think this was his point, is that capital flows 77 00:09:07.760 --> 00:09:12.280 out of the country under certain conditions. I think this is the point, I read one of Lew 78 00:09:12.280 --> 00:09:17.640 Dobbs' book, you kind of have to tease this out of the book, but I think that's his point, 79 00:09:17.640 --> 00:09:47.640 is that capital flows out of the country. And I think that's the, some people are making this outsourcing argument that we need to protect ourselves so we're not outsourcing. I think that's at least some of the more sophisticated arguments about outsourcing. I think that's their point. So I think this is the important argument. So the idea here is maybe we could construct protectionist policies that generate capital inflows. And then we would have a country that has more capital 80 00:09:47.640 --> 00:09:54.640 The idea is, if we had the capital inflows, we would have the benefits of the increased capital, there would be less trade, so we would give up some of the efficiencies of trade, but maybe there is a net benefit overall. 81 00:09:54.640 --> 00:10:02.640 So you can think of a world where, the idea is, would you rather have a country with lots of capital and not much trade, or not much capital and free trade. 82 00:10:02.640 --> 00:10:12.640 So maybe it would be better, as far as the amount of goods and services we have, maybe it would be better to have more capital, at least that is the argument. 83 00:10:12.640 --> 00:10:41.640 One of the problems here is that almost everybody who talks about this, they see the increased capital flows, but they only think of it as, they only see a partial equilibrium in the economy, they only see the protected industries and the capital inflows there, but they don't see that if we're going to import less, then we tend to export less. 84 00:10:41.640 --> 00:10:46.640 So they don't see the diminished exporting industries and they tend to not see the capital outflows in those industries. 85 00:10:46.640 --> 00:10:53.640 So they tend to not see the whole picture. That's not really a crushing argument against it, but they tend to not see the whole picture. 86 00:10:53.640 --> 00:11:00.640 A lot of these arguments, a lot of these papers were written in the last 30 years. 87 00:11:00.640 --> 00:11:11.440 But Hobbler in the 1930s spends a lot of time on this issue, and he starts this section 88 00:11:11.440 --> 00:11:18.200 about capital flows by saying, can protection generate these net increased benefits due 89 00:11:18.200 --> 00:11:22.800 to increased capital inflows? And he says, this is an exact quote, he says, definitely 90 00:11:22.800 --> 00:11:29.040 not. And then he spends several pages explaining it, and he ends up concluding that tariffs 91 00:11:29.040 --> 00:11:33.720 are unlikely to increase the income of the home population by attracting factors from 92 00:11:33.720 --> 00:11:40.720 abroad. So he seems to admit it's possible. And the two quotes are just six pages apart. 93 00:11:46.000 --> 00:11:53.000 So why are these arguments wrong here? Am I out of time, Tom? Do I have that much time? 94 00:11:59.040 --> 00:12:08.920 Then, let me say one more thing here. A second book, if you were going to read two books 95 00:12:08.920 --> 00:12:12.680 on international trade theory, I would recommend Gottfried Haveler's book. And then the second 96 00:12:12.680 --> 00:12:19.520 book is by Leland Yeager and David Turk. And it's a 1976 book called Foreign Trade and 97 00:12:19.520 --> 00:12:29.680 and U.S. Policy, The Case for Free International Trade. 98 00:12:29.680 --> 00:12:35.920 This book was, Jaeger and Turk was published in 1976, and starting in the 1980s, there 99 00:12:35.920 --> 00:12:46.600 was a whole series of papers written on optimal protectionist policies. And every time I read 100 00:12:46.600 --> 00:12:51.280 And I think I'll bet Jaeger and Turk rebutted it before any of these papers were written. 101 00:12:51.280 --> 00:12:56.440 And I can always go back to the 1976 book and say, well, there's the answer why the 102 00:12:56.440 --> 00:13:02.920 1987 paper was wrong. So it's rather an amazing, I think if you're interested in this issue, 103 00:13:02.920 --> 00:13:09.920 I think it's a must read. But Jaeger and Turk argued that if you want capital inflows, that 104 00:13:09.920 --> 00:13:15.320 free trade is the right policy because free trade is, countries that have free trade generally 105 00:13:15.320 --> 00:13:21.320 are not interventionists in general, and those are the kind of countries investors want to invest in. 106 00:13:21.320 --> 00:13:35.320 So they argue, they turn the argument around and say capital inflows are good, but protectionist policies tend to be a sign of intervention in general, and investors tend to flee those countries. 107 00:13:35.320 --> 00:13:42.320 So they argue that capital flows are due to mainly other issues other than tariffs. 108 00:13:42.320 --> 00:14:04.320 The best arguments for protection are the ones that cannot be dealt with using the law of comparative advantage, or at least that's my phrase for them. 109 00:14:04.320 --> 00:14:10.320 The reason they're wrong, or the reason I think they're wrong, is that a lot of these... 110 00:14:10.320 --> 00:14:14.400 Well, first of all, the models overlook the real world complications. 111 00:14:14.400 --> 00:14:22.840 So most of these arguments are just neoclassical model building and they start with a series 112 00:14:22.840 --> 00:14:26.180 of assumptions and then they're able to derive anything. 113 00:14:26.180 --> 00:14:32.160 So they tend to assume away the negative aspects of the protection and emphasize them. 114 00:14:32.160 --> 00:14:37.340 You can emphasize the positive aspects by saying there are certain conditions about 115 00:14:37.340 --> 00:14:43.180 about factor prices, or about price elasticities of demand for imports and exports, or you 116 00:14:43.180 --> 00:14:47.400 can claim that you know that a particular firm is a Stackelberg leader, or that the 117 00:14:47.400 --> 00:14:52.900 Cournot condition holds, or sometimes they assume that the foreign nations will never 118 00:14:52.900 --> 00:14:55.300 retaliate against your policies. 119 00:14:55.300 --> 00:14:59.300 Sometimes they assume they will retaliate, but it's in a predictable manner. 120 00:14:59.300 --> 00:15:03.700 So you know what to do about it, because you know how they're going to react. 121 00:15:03.700 --> 00:15:05.860 And a lot of times they'll assume there's just one monopolist. 122 00:15:05.860 --> 00:15:11.700 There is a monopolist, but there is only one. Because if there is more than one, then that 123 00:15:11.700 --> 00:15:16.200 throws the model off. So they are able to build models doing this, but the models tend 124 00:15:16.200 --> 00:15:23.920 to assume away the real-world complexities of the situation. However you could imagine 125 00:15:23.920 --> 00:15:30.420 a world without the modeling, you could imagine a world where the protectionism increases 126 00:15:30.420 --> 00:15:32.420 Social Product 127 00:15:32.420 --> 00:15:35.220 You can imagine it. 128 00:15:35.220 --> 00:15:38.220 You could think about a world where there's a tariff 129 00:15:38.220 --> 00:15:40.740 that attracts a large amount of capital inflows 130 00:15:40.740 --> 00:15:43.860 for industries that generate positive externalities 131 00:15:43.860 --> 00:15:47.320 and gain a large amount of monopoly rents. 132 00:15:47.320 --> 00:15:50.980 But it diminishes other industries, but those industries 133 00:15:50.980 --> 00:15:54.100 result in a small amount of capital outflows 134 00:15:54.100 --> 00:15:56.820 and those industries generate negative externalities, 135 00:15:56.820 --> 00:15:58.940 which you want them diminished anyway, 136 00:15:58.940 --> 00:16:01.860 and there's no loss of monopoly rents. 137 00:16:01.860 --> 00:16:04.820 You can at least imagine that world. 138 00:16:04.820 --> 00:16:10.380 So in that world, why can't we, 139 00:16:10.380 --> 00:16:14.900 why is free trade the right 140 00:16:14.900 --> 00:16:17.940 policy? 141 00:16:17.940 --> 00:16:18.900 Well, I think there's, 142 00:16:18.900 --> 00:16:22.940 largely speaking, there's two reasons. One is that 143 00:16:22.940 --> 00:16:26.700 in order to have an optimal tarot policy, the policy makers need to be 144 00:16:26.700 --> 00:16:35.540 be trying to implement an optimal tariff policy. And we should be suspicious of this. We should 145 00:16:35.540 --> 00:16:40.540 suspect that policy makers will tend to protect industries based on political considerations, 146 00:16:40.540 --> 00:16:47.500 not welfare issues. But to me, the main argument against it is that the policy makers in the 147 00:16:47.500 --> 00:16:53.420 real world are not able to gain the knowledge and do the calculation necessary to implement 148 00:16:53.420 --> 00:16:57.300 these policies. And a lot of these papers recognize this. A lot of these protectionist 149 00:16:57.300 --> 00:17:04.100 papers, they build a model and then they say, the conclusion of the model is that the tariff 150 00:17:04.100 --> 00:17:07.460 is the right policy under certain conditions and then they say, but you can't apply this 151 00:17:07.460 --> 00:17:12.460 model in the real world. Time after time these papers say, we just simply don't have enough 152 00:17:12.460 --> 00:17:19.820 knowledge to do it. Paul Krugman says it, and he's the leading writer, he was the leading 153 00:17:19.820 --> 00:17:34.820 This is an obvious point. None of them cite Mises or Hayek. They don't cite Mises' calculation paper or Hayek's work on knowledge, but implicitly that's sort of the argument they're making. 154 00:17:34.820 --> 00:17:52.820 But it's impossible to implement these policies in the real world, because policymakers simply, even if they were angelic beings trying to do the right thing, they simply don't have the knowledge, nor are they able to do the calculations necessary to devise an optimal tariff. 155 00:17:52.820 --> 00:17:57.820 So, Mises and Hayek's work, I think, rebut the best arguments for protection.