Inflation and exchange rate: a pastoral overview
Precious time has been bought after Brunson’s release but it isn’t clear if the said time is enough to get to the municipality elections due six months from now with a relatively cleaned-up house. I expect 5.5 as the USD/TRY goes, by the end of the year, but what matters is that the exchange rate is probably confined to the 5.5-6 box. Now, even 6 implies around 59 percent loss year-to-date, much better than of course the near 7 levels. Better still, a somewhat stable currency outlook may be formed and inflation might be ‘tamed’ so it may end up at 22 percent or so by December. Otherwise, it was bound to reach 30 percent and above. The one-off price rationing device is obviously a curiosum; I have never heard of a counter-inflationary measure that aims at cutting final goods prices by kind of an agreement with the supplying firms. Can we call it ‘price rationing’ reminiscent of the bygone centuries of pre-capitalism, guilds and all that? Well, on the face of it, this will obviously “work” for a month or two because if one indeed cuts the prices of so many goods somehow, inflation will fall, right?! For how long and at what cost are other questions. Let’s see what happens here.
Relat ve pr ces and nflat on
Consider economic theory in an abstract form. In a model with linear real assets, budget constraints are homogeneous in prices (jointly in asset and commodity prices) so prices can be defined up to a normalization. This is the usual Arrow-Debreu story. Furthermore, as in the context of Arrow-Debreu regular economies, we obtain generic local uniqueness results. In the case of nominal assets, however, there is a continuum of equilibrium allocations. In the absence of a well-defined “monetary system” such indeterminacy may occur. One implication is that the indeterminacy of equilibria translates into that of asset prices. Furthermore, when individuals are differentially informed, rational expectations equilibria à la Radner may be totally non-informative. What does all this mean? It means that if one takes a full price vector and multiplies it with 0.9 instead of 1 – a 10 percent drop in all prices - one can still obtain an equilibrium under some circumstances. This means that not only final goods’ prices but also intermediate goods and input prices should be subject to the same rescaling. If there are nominal assets, which obviously is the case in any economy, equilibrium may become indeterminate. In other words, money and credit assets should also behave in the same vein. If not, one of the following might occur: Some prices artificially drop and turnover may fall in nominal terms or EBITDAs might. If EBITDAs fall, then the taxable base will be smaller, etc. If credit and other monetary assets don’t follow, then the new ‘equilibrium’ may not obtain and even if it obtains there will be large transfers from one sector to another, to consumers but not to producers, etc. Because inflation is cost-push, the only result will be deferred inflation. Another implication is that, when the asset market is incomplete, equilibrium allocations are not invariant to changes in microlevel corporate financial policies. In the presence of secondary assets, such as options, whose payoffs depend non-linearly on the price of equity, the range of attainable re-allocations of revenue varies as the firm alters its position in the asset market. Corporate financial policy is thus relevant. When assets are nominal, monetary policy implemented through open market operations does not admit any policy-neutrality or ineffectiveness proposition. In short, such a device couldn’t work in theory.
In order for such a device to genuinely work in practice, there should have been aggregate excess demand in the previous equilibrium – which is a contradiction in terms because in equilibrium excess demand should be nil, which is not the case because demands have been falling and the economy has already been slowing down. There couldn’t have been excess demand anywhere. Otherwise, that is in the absence of excess demand, it wouldn’t make sense to infer that previous prices have been unduly high. If the price of a good is TRY 10 after the inflation pass-through and if in fact it could have been TRY 9 instead, then there must either have been aggregate excess demand or the said price should have been formed by a monopolist operating on the inelastic portion of the demand curve. This applies to the goods’ market perhaps, or to ‘real assets’ for that matter. For nominal assets, the same logic should be valid also. If all relative prices will fall by 10 percent, including not only the price of goods and inputs but also the interest rate, etc., then obviously we will have a new equilibrium because general equilibrium admits normalization in the canonical version. To me, this might only be called, rather benevolently,
a device aiming at rendering inflation ‘tame’ in the short run at the expense of the longer run unless previous prices have been ‘wrong’ somehow and costs didn’t rise that much, or else all relative prices should fall at the same rate.
Exchange rates and nflat on
Hence enters the exchange rate anew. It isn’t solely the exchange rate that is responsible for the rise in inflation. However, in the short run it is the main driver. Should the USD/TRY exchange rate remain between 5.5-6 within the next quarter, then the year-end inflation will be less than anticipated a few days ago. I was forecasting at least 30 percent CPI in the coming months as the peak level because producer prices are higher than consumer prices by 22 percentage points, because producer prices and the core CPI rise in tandem with near-correlation with the exchange rates, because oil prices might go up further, etc. Now things may change a bit. We call it the ‘pastoral overview’ driven by the exchange rate. None of the resilient economic problems that have led to the current stagflation-like panorama have disappeared, of course.
We had better note that this isn’t the wake of a balance of payments crisis. It has to do more with the debt and revenue structure of the corporate sector mostly. Turkey imports almost everything and Turkish firms are heavily FX-indebted. Furthermore, they carry a sizeable short position, i.e. $216 billion corresponding to 27 percent of GDP. With lira depreciation, this ratio will climb higher, and fast. No, higher ex ante real rates of interest alone will not suffice. Because inflation is now bordering the old “moderate inflation” level, i.e. in the range of 20 percent, and because such an inflation can go either way, no ex ante real rate provides a guarantee in itself as a means to defend the currency. A hike was necessary because the CBRT is well behind the curve, but that’s about it. A sizeable hike was a necessary condition, not a sufficient one. Even now, the secondary market benchmark TRY Treasury bill rate is two percentage points higher than the CBRT policy rate. I don’t expect any further move from the CBRT in the coming month because the exchange rate might stabilize now. If inflation expectations fall either because the Brunson problem is now over or because the 10 percent price cuts might ‘work’ in the short run to some extent, then the ex ante real rate will rise and commercial loan rates might stick to their current level, or even fall a bit also. Nonetheless, in our basic view of the world, syndication costs will rise further in 2019 due to the Fed and ECB balance sheet and interest rate hike effects.
Interest rates, growth and nflat on
Will the Fed continue raising interest rates for at least a year? The answer is: more to come but not fast. The Fed will take into consideration the possibly negative impact of trade wars on the U.S. economy and the likelihood of EM meltdown. Hence, interest rates will be raised sequentially. Since the Fed is independent, whatever message conveyed by Mr. Trump is ignored as we have seen once more. Pen in 1 + 3 hikes until end-2019, which means 3.25. Since there are at least two more urgent problems here, the unstoppable change in the direction of Fed monetary policy comes second now. Nevertheless, I think syndications will be rolled-over at around 100 percent, albeit with an at least 1.5 percentage points (visible) cost increase. At that point, the Fed balance sheet change and rate hikes will become a key variable again even among local plays here. The other problem, i.e. private sector’s FX debt, is waiting for a solution. The idea that the U.S. yield curve reversion, should it occur, would bring the Fed’s march to a halt seems unfounded now, especially given that the U.S. budget is in deficit – $900 billion until the end of the year - and the Fed is still selling securities at a speed of $50 billion a month. So, the U.S. has to incur debt anew, and the market rates are still too low to be equilibrium rates. Remember that the 10-year U.S. Treasuries were carrying 4.3 percent right after Lehman. Pen in at least 50-75 basis points additional increases on financing costs for 2019, and not only for dollars. The Eurozone is still to come, possibly in late 2019.
Recession is a technical term. By the most common and simple definition, it is a consecutive two quarters contraction of the GDP. The Turkish economic malaise has many roots, and not only the construction sector, for instance. However, the situation is rather new, although anticipated. There exists now also a construction-cum-real estate cycle, and this is a novelty. True, the Turkish ‘mortgage’, if one can use the term for less than 10-year housing loans, is nothing like the American mortgage, which comes quasi-automatically with a 30-year maturity. Therefore, the American mortgage carries a ‘home equity’ tag, refinancing facilities, recourse and non-recourse lending, syndication of credit tranches, MBSs, CDOs and what not. Not here, and it is a good thing. Nevertheless, a good deal of problems come from excess supply of homes in various cities, financing difficulties some or many developers currently face, the fact the sector is heavily dependent on imports – hence, the steep rise in construction costs that has already occurred in the last couple of months, and backward-forward linkages deeply entrenched in the economy at large –which incidentally implies an important negative overall impact. Sectoral employment has already fallen by 145,000 since the beginning of the year, and can only be expected to nosedive in the near future. The panorama looks brighter, but that’s about it. Fundamental problems remain. Good start though: the exchange rate now matters much because inflation is pervasive and inflation expectations ought to fall so we can form for early 2019 an exchange rate-cum-interest rate-cum inflation mix forecast that looks stable.