What might the Eurozone crisis have looked like without the euro?

(Originally published on Tumblr, March 10, 2014. Just trying to consolidate blogs with some past greatest hits. Apologies if you have seen it already. I do plan on doing new posts on the stack).

Obligatory Dad Joke Starter -- Bismarck's revolution from above is a classic example of the state's relative Ottonomy.  




This piece is in part a reaction to the recent hullabaloo about emerging markets facing a crisis that is comparable in severity to the 1997 despite having moved from fixed to floating exchange rates. Although I don’t agree with this view, it did get me thinking about the way completely open capital accounts affect real exchange rate developments, investment decisions, and the broader economic cycle. It seemed to me that based on this, one might make a plausible case that the problems of the Eurozone might be less the result of an inflexible currency regime and more the result of the Treaty-enshrined commitment to the free movement of capital inside the EU, regardless of whether the country in question is in EMU, the EMU ante-chamber or has a formal (or de facto as in the case of Sweden) opt-out.

My conclusion is based on a thought experiment. Imagine that there had been no monetary union in 1999 but that the EU did contain a number of different countries at different levels of per-capita income that were undergoing a process of real economic catchup and convergence. By virtue of their membership in the EU, they would have been formally committed to the free movement of capital even while they retained independent central banks and their own national currencies. Under these circumstances, at the beginning of the 2000s, one might have expected very low German interest rates and a weak Deutsche Mark and a strong Peseta and high Spanish interest rates. It is possible that those high Spanish interest rates might have restrained the construction boom that followed, but it is just as likely (and arguably more likely) that what we might have seen instead is a massive flow of hot money from Germany to Spain in the form of a carry trade seeking to capitalize on the interest rate differential, which would have tended to loosen Spanish monetary conditions. Concurrently, we would arguably have seen a large increase in DEM denominated borrowing by Spanish entities.

Either of the above would presumably have led to an even stronger Peseta in nominal terms, the effect of which would have been to discourage investment in the tradeable sector (due to currency overvaluation) and instead funnel that investment into the non-tradeable sector . Eventually, at some point, we would likely end up with boom turning to bust, potentially large asset-liability currency mismatches inside the banking sector, a supply overhang in non-tradeable investment, and a much weaker currency. However, it is important to note that the benefits of a weak currency in the bust might be limited precisely because of the focus on non-tradeable investment during the boom years when the currency was too strong. If anything, the pace of nominal appreciation under those circumstances would probably have militated even more strongly against using a surfeit of foreign capital to invest in the tradeable sector precisely because businesses’ investment decisions are likely driven to a greater degree by nominal appreciation than by real appreciation due to higher non-tradeable inflation (which is what actually occurred).

I think it is reasonable to suggest that my counterfactual for Spain would have turned out the way I suggest above, precisely because we know that is how it turned out in the UK, Iceland and Hungary, all of which had floating exchange rates. In other words, the actual outcome for a country like Spain might not have been terribly different whether or not it had retained the Peseta. There is a question of magnitudes at least during the boom years, which depends in part on whether capital inflows driven by the seeming security of the fixed exchange rate were in fact greater than carry/appreciation-seeking inflows would have been under a floating Peseta regime. I don’t know the answer to that, but my gut feeling again is that it wouldn’t have made that much difference. What I do know is that in countries like Australia and New Zealand in 2007, the combination that central bankers desired most ardently was higher interest rates and a weaker currency (in order to both raise domestic savings and to rebalance the economy towards exports away from consumption or housing) but despaired at being able to bring that about. Similarly, during those years, the market had no compunction about funding current account blowouts on a scale comparable to or in excess of Spain’s imbalances (in Iceland). 

Another way to think about it is to ask the question of whether those famed unit labor cost divergence charts would have looked better or worse in the event that the countries of the EU had retained both completely open capital accounts and their own currencies. Again, I have no idea, but my gut instinct is to think things would have looked even worse given the tendency of markets to push nominal currency overshoots. In other words, nominal appreciation would have led to even more ULC dispersion than the different rates of non-tradeable inflation that lead to the same result in those charts. In other words, I think there’s a decent case that the boom years would have been even more extreme in Spain had it retained the Peseta.

What about the bust? How might the retention of national currencies might have made a difference?. In this regard, I think one key is whether national banking systems would have (in the counterfactual) accessed cheap capital in their own currencies or in DEM or USD. I suspect there is a strong likelihood that large chunks of the private sector would have just gone for the carry trade/lower rates and indebted themselves in foreign currency. This in turn might have limited the ability of national central banks to deliver countercyclical monetary policy once the bust came because they might have needed to limit depreciation pressure on the currency. After all, by construction, the third leg of the Trilemma (capital controls) is formally closed to EU members. 

Furthermore, by construction, the absence of a currency union would have meant no Target 2—i.e., no ability to transfer the currency pressure from herding private sector exit to the public sector (which, in my opinion, is at the heart of the Buba/periphery Target 2 imbalance problem). At its peak, one can imagine what the equivalent of a 700 billion Euro exit from peripheral assets to the core would have done to core/peiphery bilateral exchange rates, and the corresponding impact on private sector solvency. The nominal exchange rate overshoots to the strong side during the boom years would likely have been dwarfed by overshoots to the weak side during the bust that followed.

However, even in the event that a large portion of the boom era debt had been contracted in local currency, what this would have done is to transfer a portion of the problem to the creditor countries. Essentially, one would have had core banks facing a sharp drop in the DEM/NLG value of overseas assets due to currency depreciation. Furthermore, one can imagine that an Ireland that did not need ELA (since it would have had its own central bank) might have had fewer compunctions about imposing losses on the creditor countries in the wake of the Anglo-Irish debacle, with even more disastrous spillovers into the entire European (and probably global) wholesale bank funding market.

As to the real economic effects, whether the core countries faced the default of their overseas borrowers indebted in core currencies or losses in DEM terms due to massive currency deprecation of their peripheral assets, in either instance we would have wealth losses at the core. This would have had a corresponding impact on domestic demand, easier monetary policy, currency depreciation and efforts to rely on external demand and larger current account surpluses at the core. But precisely because of the nature of investment (predominantly in non-tradeables) undertaken in the periphery, they would have found it extraordinarily to difficult to gain market share from the core even with weaker currencies.  

Indeed, I think it very likely that the chain of depreciations in both core and periphery in my counterfactural scenario could have raised pressures for protectionism, capital controls and threatened the entire edifice of the common market. Rather than exporting deflation, the EU might be blamed for spreading financial distress, capital controls and competitive devaluations around the globe. Instead of charts comparing unemployment levels with the 1930s, we would probably all be looking at the famous Kindleberger chart showing global trade in a declining spiral through the early 1930s. There might have been some mitigating factor in my alternative scenario stemming from the absence of the Stability and Growth Pact (and consequently, more space for a fiscal response), but I’m unconvinced that the market would have tolerated countercyclical fiscal policy from countries undergoing a currency blowup on the scale described above.

What would probably be different if we did still have European national currencies would be distribution of losses between core and periphery and between losses to wealth and losses in employment. Nevertheless, at least to me, the counterfactual seems sufficiently awful that might all be wondering whether it made sense for a series of small, open, highly-interlinked economies that were extremely porous to foreign capital and with tightly linked production chains to have national floating exchange rates and national financial regulation. In other words, we might be wondering whether we would not be better of with a common currency, a common central bank, and a common system of banking oversight, and a genuine financial market union.

Perhaps I am wrong (and I suspect that I might be more wrong about the bust phase than the boom), but the the purpose of the above thought experiment is just to raise some questions about the catechism that everything is fine with free markets as long as you float. And it comes from someone who used to believe in that catechism. That is because I think there is increasing evidence that in world where currencies are driven less by trend shifts in trade balances and much more by very large, quasi-instantaneous shifts in desired portfolio balances, there is a good reason to believe that floating exchange rates can function as shock enhancers rather than as shock absorbers.