Economic analysis of policy proposals may be based either on first principles or on economic modelling. The proposed FTA is too complex to be analysed simply in terms of first principles. Nevertheless, a great deal of insight can be obtained from simple parametric models of various aspects of the proposal.
As compared to a large-scale simulation model, this approach has the advantage of clarifying the processes leading to estimates of costs and benefits. A large-scale model offers greater precision and the capacity to model policy outcomes for particular regions and industries. However, where there is a large divergence in estimates of aggregate outcomes between simple and elaborate models, this divergence is rarely a consequence of greater precision in the elaborate model. More frequently, the divergence is the result of differences between the economic assumptions used to ‘close’ (that is, derive an equilibrium for) the elaborate model and the economic assumptions used in the simple model. Hence, there should be no automatic preference for the results of more elaborate models. What matters is the validity of the core assumptions.
John Quiggin 
John, I am not a trade economist but is the following roughly true?
For a small open economy like Australia there is a strong efficiency case for moving market-by-market towards freer trade irrespective of the trade policy stance of other countries. This is so for any region in Australia that can trade more freely in goods or assets with any region elsewhere in the world and works if the reforms are piecemeal or if they are all-inclusive. You can reject efficiency as the objective and use other arguments against free trade on, for example, distribution issues or adjustment costs but these are other stories.
If Australia does have real bargaining power in trade negotiations and sees freeing up trade as a concession to the rest-of-the-world rather than something that unilaterally benefits us then the issue is more complex.
So do we have much bargaining power or not? This might not be indicated by the ‘difficulty’ of the bargaining process since the ‘other side’ has political and other incentives to make a virtue of what it knows is in its own self-interest.
If we don’t have much bargaining power, and it is difficult to understand why (ignoring perhaps military paybacks) we should in relation to the US, then even partial liberalisations seem to always move us in the right direction and simple qualitative models calibrate gains correctly. More elaborate models catch spillover effects and perhaps give us more accurate estimates but will not improve our ability to steer the ship in the right direction.
I think it’s more complex than HC supposes. He mentioned spillovers, and I do see those as able to cause problems. But I do see another problem area as well, ownership, in particular ownership of the gains.
It is quite possible for more than 100% of the gains to go elsewhere, so that even though the “right” business decisions are being made, Australia ends up worse off. Obviously spillovers are only one way of ownership ending up in the wrong place; indeed curing them through privatising can give them to outsiders without value in exchange if they get the new revenue yielding assets at a discount.
In theory, outside investment puts in compensating resources and balances things out, so Australia would still be a net gainer even with outside owners. But the privatising at a discount example shows that this is not necessarily so, and what is more there are currency movement things that mean outside “investment” might only be nominal and not a real transfer of resources in exchange. This actually happened under neocolonialism, and it is probably happening now from the US dollar being both a depreciating fiat currency and a reserve currency – printed money flows through to acquiring assets for nothing real. (Though it is probably middlemen countries that are buying our farm, not the USA.)
The harm is exaggerated by our being a primary producer; what we export drives value chains outside our economy. If we exported high value products we wouldn’t suffer so much from absentee landlords (see Nassau Senior comparing England and Ireland in this respect).
So I see “free trade” as being dangerous even when it does the “right” thing from an efficiency point of view, since we can easily get less than 0% of the gains, i.e. lose.
Of course I’ll defend models… I’m one of the guys doing the modelling for the US FTA (that’s why I’m at work on Sunday). But I agree that complex models aren’t necessarily better models. The value of models is (among other things) that they help you isolate what is and isn’t an important assumption.
Harry – preferential trade liberalisation can create a net loss in allocative efficiency because we are treating the same good differently dependent on which country it is made. Note that I only said can… to know whether it does or not you have to do the modelling (in other words… I’ll tell you in a week).
It’s easy to get rid of this potential negative (referred to as ‘trade diversion’) – you simply make your tariff reductions MFN (which means, you cut tariffs on everybody).
As for whether Australia will get the gains – if it’s allocative efficiency gains, yes. If it’s dynamic efficiency gains, yes. If it’s a gain through increased investment, yes. If it’s a gain through better access to US government procurement, yes. So fear not PML.
Yes PM, foreign ownership can have perverse effects. For example if land is an abundant factor and, because of a tariff cut, we can better export agricultural products then returns to land would improve while wages and returns to other scarce factors deteriorate. So if the land had been sold by Aussies to foreigners prior to knowing the tariff cut was to occur it would have been sold at less than its value and Aussies would be worse off as a result of freer trade.
Yes its possible. But there are a lot of ‘ifs’ here. Generally if I want to buy and sell with you and any barrier to our trading is removed (tariff, transport cost) we are both better-off. And it was this simple view that was driving my argument for having faith in simple models.
JM your ‘trade diversion’ argument puzzles me and my forgotten trade theory might be the reason. Mr A has a chance to trade with two others (Mr B & Mr C). High taxes limit both trades. One tax (an import tariff imposed on sales from B) is reduced but the same high tariff persists on trade with C. Surely I am never worse-off as a result of the cut? I might be worse off than with taxes lowered for both and I might have been better off had the same reduction occurred in trade with C rather than B but surely I am never worse off than I would be in my original situation where both taxes were high. Or do I misunderstand?
Harry
If Mr B is a less efficient producer than Mr C, then you might be worse off, if trade is diverted from C to B. You’ll get your goods from B tariff-free , but still maybe at a higher price than if you’d got them from the efficient producer C, even with a tariff.
Uncle Milton. If B is less efficient than C but both are subject to protection I am still not worse off if tariffs are reduced in B? You say I would have been better-off sourcing the goods from C even with the tariff. But if that was true I would buy from C and the tariff reduction in B would not affect the source of purchases. I wouldn’t switch to B but would be no worse-off. You might say that it would have been better to initially cut tariffs with C but this was initially agreed to. I still can’t see why it is not true that a sequence of trade liberalisations in whatever order and whether partial or not always improves my (A’s) welfare. Hence recourse to simple models make’s sense.
Geographically: I want to get from X to Y and a series of barriers are in my path. If any barrier is removed my effort in getting to Y is reduced. If another barrier is reduced my effort is again reduced. This is different from saying that the barriers removed were ‘best’ in terms of reducing my effort.
My trade theory is also a little rusty (I’ll be brushing it up soon), but I think Uncle Milton is on the right side of the argument here. I certainly recall results saying losses from trade diversion can outweigh gains from trade creation.
However, I think the loss is most obvious in relation to B. By hypothesis B has a comparative disadvantage in the world market for the product in question, but the preferential removal of tariffs by A leads B to expand output. It seems plausible that this can imply a welfare loss for A and B combined. It’s less clear how the loss is distributed between A and B.
John, I don’t see how. A reduction in protection between A and B must make both A and B not worse-off. Hence A+B must be not worse-off. I have mainly argued that A alone must be not worse-off. It might be true that A+B+C (the world) are worse-off because production could have much more cheaply occurred in C and C could be heavily damaged by the trade diversion but this is a different issue — if that’s what is meant I do (finally) understand the ‘trade diversion’ argument.
Harry, the trade creation – trade diversion argument goes something like this. Initially, I am buying from C (the more efficient producer) rather than B. I enter into a trade agreement with B where I buy from B tariff free but which locks C out of our market with a common external tariff.
There is both trade creation (trade between me and B) and trade diversion (away from the efficient producer B). Welfare could go either way.
That is the basic model. However, I think the way this can be reconciled with your intuition is if C isn’t actually locked out of my market. If I can continue to import from C at the old prices, then I agree I will be better off, or no worse off.
I think this comes down to the basic distinction between customs unions, which have a common external tariff, and free trade agreements, which do not.
However, actual FTAs may in effect be just like customs unions depending on things like rules of origin. So, let’s say I want to export a shirt to the US under the USAUS FTA. Right now, it’s made of cheap Indonesian textile. Under the rules of origin of the FTA, I can only get it in tariff free if it’s made with expensive US textile. As a producer I might or might not be better off. I sell more shirts, but the production costs are higher. US consumers might or might not be better off. They are getting tariff free shirts, but using inefficient US resources to make them.
However, if under the FTA I have the option of exporting shirts under the old rules, with high tariffs but cheaper production, then I think I can be no worse off, but you’d have to ask a trade expert to be sure.
But what tariffs were reduced by the US in the FTA that will have any signficant affect on Australian export trades? – none in the area of agriculture AFAIK. Reductions in manufacturing tariffs are more likely to impact negatively on Oz manufacturers than the other way round, especially with a falling US$. Sure this is positive in terms of allocative efficiency but isn’t this a bit like GW Bush arguing, to howls of derision, that out-sourcing is good for the US economy whilst employment stagnates(last months growth notwithstanding)? Doesn’t this beg the question of all economies heading for the lowest common denominator – the price of prison labour in China? As someone pointed out to me – a $20 angle grinder from W.A. Salvage could only be made in this fashion!
There are two other factors that matter here apart from tariffs. The price of fuel/transport is probably at least as important and relevant in terms of bulk commodities versus high value products. So tariff reduction doesn’t matter as much for agriculture (I know I’m contradicting what I said above)
The other is the question of dependency. PML mentioned the issue of ownership which is a special case in a way. All trade creates a dependency (or interdependency if you like) on another market and conditions that prevail there. This could be seen as increasing the risk or as spreading the risk, but only the latter if a local industry still survives.
To give an example, dairy industry rationalization has spelt the death knell for the W.A. dairy herd, at least far as milk production is concerned. The dairy farms that have not already gone out of production are living on borrowed capital and borrowed time. Some will reallocate their resources by going into cheese or other kinds of production but given that the value of their stock has decreased the capacity to borrow capital to diversify is reduced so most will simply go broke. This process is being exacarbated by the aggressive pricing of imported UHT milk in W.A. supermarkets.
The result is that W.A., in a few years, will not have any capacity to supply milk to itself. Consequently when the price of fuel goes up (when not if!) and local competition is eliminated the price of milk in W.A. will increase rapidly – fuel prices will most likely be used as an excuse for profit gouging by the importers.
The real problem that modellers face in this situation is that reproductive goods are ‘sticky’ in terms of cost. Hence you get boom and bust cycles built in and exaggerated by a reliance on models that regard all goods as products rather than some goods being living things.
Thanks Uncle Milton, I understand. But you are right I had a different picture in mind.
So according to your view initially A trades with low cost producer C but switches to high cost producer B with an FTA and a common external tariff directed against C. Then A can be worse off — I have no disagreement.
I was considering the case where a high external tariff was not subsequently imposed. Then I claim both A and B are no worse-off although A+B+C can be worse-off.
Harry, if the US FTA is truly MFN, then it appears on first principles that Australia can only win from the trade liberalization, though how much is another question.
Of course, since FTA appears to lock us in to the US intellectual property regime, which is increases the monopoly rents to IP owners, then the net welfare position could still be negative, even under MFN.
By the way, John Humphreys, how is it that you know that there will be allocative efficiency gains, dynamic efficiency gains and all the other gains, when you won’t have finished the modelling for another week? Aren’t you supposed to wait until you see what the model comes up with before you draw your conclusions?
Surely, it cannot be true that you are simply going to give your client, DFAT, the answers they want to hear.
Uncle Milton,
Professor Peter Dixon told me at our annual La Trobe/Monash Uni cricket match (we lost again!) that he thought the net benefits to Australian from the FTA might be small or negative after accounting for IP issues. Increasing enforcement of property rights restricts trades and can be immiserising even if the vendor is not a monopolist.
John Humphreys wrote “…If it’s a gain through increased investment, yes. If it’s a gain through better access to US government procurement, yes…”
The thing is, I know from economic history that it ain’t necessarily so, so we really do have to worry about the possibility and look into whether it is happening (or if people are contemplating it). The funny money to acquire resources trick was used by the French in Italy and the low countries in the revolutionary period and in North Africa under “peaceful penetration” (though that was more tax driven than using funny money). It was used on Vichy France by the Nazis, and through them on North Africa. Between those periods it was used by the Dutch in the East Indies, when they depreciated the currency to finance setting up the “Culture System” of exploitation. It doesn’t show up much in British colonial history, since that didn’t often penetrate more developed areas after the tricks were invented (India was earlier, and Egypt was actually penetrated by the French and only transferred to the British sphere).
So I know the trick and how it works: print money and acquire assets to get hold of a long term yield. Since this is a wealth transfer, you simply don’t have “increased investment” – I do know that investment does make ownership changes give value all round, it’s just that they don’t have to happen. And within that context “better access to US government procurement” only means that those procuring get the benefit – they are functioning as middlemen and facilitating the wealth transfer from their compatriots to outsiders. (In this case, the most likely gainers are not the USA but the USA’s trade partners trading in high value added items – which is one reason why they continue to accept US dollars, since they will gain at yet others’ expense.)
I would have used the example of Zionist funds entering the Palestinian Mandate, only it would probably cause more heat than light. However it’s worth noticing that inflation there was only headed off by such things as the (defaltionary) commutation of tithes, a British reform.
*** Just to catch your attention, let me point out that you’re all wrong. 🙂 Trade diversion occurs when you buy from somebody who is not the least-cost supplier. Let me be clear – the person buying the product is not worse off! However, if you’re switching to a higher-cost producer, then the gain to the consumer is exceeded by the loss of tariff revenue. Therefore, the country as a whole can be worse off!