Monday, August 31, 2009

Weekly Hurriyet Column: The devil wears a headscarf

Below is the unedited version of my column for this week. You should be able read the final version at the Daily News website, but again, I could not find it (Gul, I am thinking about disclosing your email and cell number so that my loyal readers can take it up with you), so if you can, please email me the link:)

Note that the government's own projections in the Medium-Term Economic Program (MTEP), which were released two weeks after I wrote this column, are more or less in line with mine, with the important exception of inflation, which I outline below in great detail. In fact, with the government projecting a budget deficit of 6.6% of GDP, it has taken my "number of the devil" pun further, so now noone can accuse me of non-PC behavior...

Unfortunately, despite the sound forecasts, the MTEP is far from quelling the devil, mainly from the weakness of its fiscal framework. I will detail what I mean in the preface to next week's column, which deals with the fiscal framework.


While updating his key economic forecasts in June, your friendly neighborhood economist stumbled upon an interesting observation.

Turkey’s holy trinity of economic forecasts was yielding the number of the beast, with my 2009 growth and budget deficit forecasts coming at -6.66 percent and 66.6 billion liras and end-year inflation expectation at 6.66 percent. After more than two months and a plethora of new data releases, it is time to go over these projections again. I take on growth and inflation today, leaving fiscal policy to next week.

The dismal first quarter GDP figures and the more recent data have, if anything, confirmed my dismal growth outlook. While it is true that private consumption and investment have bottomed out, media and economists alike seem to have been misled by the fiscal stimuli in the second quarter. With the consumption, real sector and confidence indices all having fallen recently, all we can hope for at this point is a tame W, with a small inflection point.

In the perennial optimist’s Turkish economy story, much ado is being made about nothing, i.e. Turkey’s much-hyped-about non-leveraged consumers and well-capitalized banks. However, the former will not splurge and the latter not lend unless a coherent economic framework is in place, as the latest data attest to. Even then, tighter external financing and a permanently-higher unemployment plateau will slow the speed of the recovery. I see yearly growth contraction at around 8 percent in the second quarter, hopefully with some positive surprises from housing and agriculture, and somewhat less in the third. Then, a modest positive growth in the last quarter will take us to a yearly figure of around 6.5 percent.

Unlike my growth outlook, my inflation projections need some updating, as it now seems that the country’s large output gap is here to stay for a while. Moreover, the effects of the considerable slackness in the economy, lower oil prices and the tax cuts have been larger than I had envisaged. As a result, I now see end-year inflation at 6-6.5 percent, which is still above market expectations of just below 6 percent. However, I believe I have only miscalculated the speed of the wagon, not its direction. I still see inflation heading uphill from here- in contrast to the Central Bank’s (CBT) downhill projection.

For one thing, the fiscal disarray, which I will outline in next week’s column, is bound to lead to a sharp pass-through from the higher oil prices in the coming months. Moreover, I believe that the recent weakening of the link between inflation and exchange rates is, contrary to what many economists claim, only temporary. Another misconception is attributing the recent taming of durable goods inflation solely to the output gap without acknowledging the contribution from tax cuts and inventory drawdowns, which are already over.

Once producers get some pricing power, probably during the last quarter of the year, exchange rate pass-through will return with a vengeance. Further complicating this gloomy picture are inflation expectations, which have proved to be quite sticky. The devastating effect expectations could have on services inflation over the medium-run is overlooked by many economists, partly owing to the CBT’s downplaying of this mechanism. As a result, I see inflation heading north rather than south in 2010. All this means that despite CBT’s assurances in its latest Inflation Report, policy rates will probably have to be moderately hiked next year.

My number of the beast forecasts have not changed by much. But at least we have just the right government to quell the devil; hopefully, it will be with the much-awaited economic framework.

Sunday, August 30, 2009

Speacial Hurriyet column: An industrious look at industrial enterprises

In addition for the weekly columns, I also do special analysis pieces from Hurriyet from time to time. I wanted to do an analytical piece on the Istanbul Chamber of Commerce (ISO) Top 500 Industrial Enterprises survey when it came out in July, but then decided to wait on for the second-500 list to be able to work with more data. I did the analysis with Ahmet Asarkaya, a very able ex-student from my teaching days at Izmir, who now happens to work in Turkey Data Monitor.

Anyway, as is the norm, below is the unedited version; you can read the final version at the Daily News website. One big difference between the two is that since the editors decided it, unlike my weekly would not be an opinion piece, the first person narrative had to be transformed to third person.

I am big fan of of the maxim "a picture is worth more than a thousand words", so below are the figures that go with the piece; the paper could only publish the second one due to space limitations.

Last but not the least, I learned by chance that ISO only publishes some of the data they collect; in fact, they get detailed balance sheet info, which would make their data the best panel data on Turkish companies. I am in talks with them to use the data for research, and if they agree, I and Ahmet plan to do more, especially interesting panel regressions. So you'll probably hear more on ISO data from me soon.




On Wednesday, Istanbul Chamber of Commerce released the results of its Turkey’s Next Top 500 Industrial Enterprises 2008 Survey.

The survey and its big brother covering the first 500 firms, which was made public a month ago, make up one of the best panel datasets of Turkish firms, so we had always been a bit disappointed by the indifference of academics and media alike towards them. While we have yet to see serious academic work making use of the surveys, the crisis has brought unprecedented media attention. Late to the game, we would like to highlight some of the uncharted results and dispel some chronic myths.

One of the recurring themes of the data is that big is beautiful. The top 200 firms are more than five times more productive (in terms of value added per worker) and profitable (as measured by return on assets, or ROA) than the bottom 200. More surprisingly, these much-publicized findings seem to be a result of aggregation more than anything. There is hardly any correlation between size and productivity or profitability, and whatever there is completely disappears once you account for sectoral differences.

Big is not only beautiful, it is also lean and mean: While the main take of the survey by the press, that Turkish industry has been devastated by the crisis, is unfortunately true, as depicted by the record decrease in profits (16.5%) and increase in losses (150.4%), larger firms have been able to relatively shield themselves (at least compared to their smaller counterparts), not only in this crisis, but also in the 2001-2002 episode. To illustrate, while the top 200 saw its profit margin decline by 2.6% from 2007 to 2008, the same figure is 3.7% for the bottom 200.

The dismal state of the smaller firms would be a just cause for concern, as they are arguably much more representative of the country’s small and medium-sized enterprises (SMEs) than the powerhouses in the top 500. But interestingly enough, it is the middle-pack that is displaying the largest declines in some negative performance scales. This surprising result is actually in line with previous findings, notably in the World Bank Investment Climate Assessment Survey of 2005, which found out that middle-sized lack the muscles of the behemoths and the flexibility of the SMEs. Then, maybe, big is beautiful, only if you are big enough.

One group that should have been affected disproportionately from the crisis is exporters: After all, with Turkey’s main export markets hit hard by the crisis and export volumes having registered sharp falls after the first quarter of 2008, one would expect exporters to be utterly devastated by the crisis. In fact, they are anything but, at least the ones on the ISO list, who have been able to maintain their ROAs and actually increase their profit margins. While deeper analysis is needed, we suspect a selection bias is going on, with the ISO list featuring crème de la crème of the exporters; the ones that have been able to rapidly shift their markets.

In a similar fashion, foreign-owned firms, as defined by non-domestics owning more than 50% of the firm, have been able to shield themselves better from the crisis as well. In general, these firms are also more productive than their domestic-own counterparts. But correlation should not be confused with causation: It could also be that better-managed firms attract foreign capital. However, a closer look leads us to partially reject this hypothesis, as foreign-owned firms are not that different from their domestic counterparts before they are acquired. Foreigners do not work their magic immediately, either: We found no difference in terms of profitability and productivity in the couple of years before and after a firm is acquired. The benefits seem to come later on.

There are nevertheless important sectoral differences as well. For one thing, all but two sectors (mining and electricity/energy) have seen their profit margin and ROA fall from 2007 to 2008, with the sharpest falls being in paper products (which include the newspapers to our dismay), textiles and food & tobacco. The latter is one hell of a surprise, as it has been touted in the media as rather unscathed from the crisis.

One important disclaimer of all this analysis is the timeframe of the survey. Since last year, we have seen a record-level growth collapse in the first quarter. But we have also witnessed temporary stimulus measures in certain sectors resulting in a likely quarter-on-quarter positive growth in the second quarter. It will be interesting to see how these factors have played out in the top-1000 industrial enterprises this year.

One of the authors of this study painted a rather contrarian view of the tourism sector in this paper titled For whom the bells toll? some time ago. He now completely regrets for having already used this title, which belongs more with industry. Compared to industry, tourism is in one great party.

Monday, August 24, 2009

Weekly Hurriyet Column: Realism and its evil sisters

Below is the unedited version of my column for this week. You should be able read the final version at the Daily News website, but I could not find it (Gul, take notice please), so if you can, please email me the link:)

Since I wrote this column four weeks ago, the benchmark has stayed more or less in the 9-9.5% range, despite a very favorable global environment and the CBT cutting rates 50bp, as expected. But now, it seems that the CBT may not stop with a final 25bp cut next month. While the ambiguous tone in the one-pager accompanying the rate decision led to a never-seen-before divergence of opinion between analysts, most think that CBT will take rates to as much as 6.50% from the current 7.25%, meaning that an additional 50bp (than I envisaged in the column) may be in the cards. But the analysis I did for the column shows that this should not have a huge impact on benchmarks rates, certainly less than one for one.



Last week exposed me to realism and its evil sisters, unrealism and surrealism.

I find the latest outcries by some economists for the Central Bank (CBT) to take the policy rate to 5 percent more surreal than a Dali painting. After the Bank’s latest expectations survey, the ex-ante real policy rate now stands at 1.4 percent, a historic low for Turkey. But rates are zero, or even negative, not only in developed countries, but in emerging market peers with flexible exchange rates and inflation targeting such as the Czech Republic and South Africa (a notable exception in Brazil, which I incidentally see as closest to Turkey).

Such naïve comparisons, however, ignore country risk. Adjusting for that with credit default swaps (CDSs) reveals that Turkish policy rates do not have much more room to go, maybe another 0.5-0.75 percent or so, which is in line with expectations. And that is after noting that Turkish CDSs are low compared to the country's credit rating. But this doesn't mean much, as the low-interest camp argues that Turkey's credit rating is too harsh to begin with, an argument I would agree, if I didn't know the dismal fiscal outlook.

Assuming a 0.5-0.75 percent cut in the policy rate over the course of the next two months, after which the CBT will stay put for a while, the next step is to answer the two billion-dollar questions that are very much interrelated: How much will the rate cuts pass on to market rates? When will this aggressive easing start to affect the real economy?

To answer the first question, a variety of approaches could be employed: For example, the difference between rate expectations today and a year from now gives a rough idea on where the benchmark is headed. This approach yields a low of 9.2 percent. Calculating expectations from cross currency swaps yields a similar minimum benchmark rate. Alternatively, equating funding costs to the duration of the benchmark, or looking at yield curve expectations 3 and 12 months down the road, I end up with 9 percent.

It is possible to integrate all of these approaches into an econometric framework, taking into account global developments as well. This comprehensive approach, which takes me to just below 9 percent under a very optimistic scenario, also demonstrates that the marginal effect on the benchmark from taking the rate to below 7 percent would becomes less and less. In sum, there is some more room for policy rates, but not that as far as the 5 percent some are recommending.

Coming to the second question, if it were left to the media, credit markets have not just started to thaw, they are liquid water already. Notwithstanding the difficulty in discerning how much of the recent easing in certain credit rates is actually due to CBT actions rather than to the global easing of credit, as demonstrated by global indicators or Turkish banks’ recent ability to borrow from abroad, there is hardly a stir on the data front, which is lagging behind just a couple of weeks.

Deeper analysis darkens this rosy outlook further. Adopting a methodology developed by Roberto Rigobon of MIT, it is possible to discern demand and supply from quantity and price (interest rate) data. Although I am at early stages, results so far attribute the credit freeze to a lack of demand as much as supply. While this partially exonerates banks from countless accusations, it is a bit unrealistic to assume that lower rates will induce consumers and businesses to borrow in an environment of uncertainty even if banks are willing to lend.

I find it very amusing to see that the ultra-low real rates have led to an economic landscape of unrealism and surrealism, where optimism has replaced looking into the facts and data.

Sunday, August 23, 2009

More on Treasuries

I have been burning some brain cells on where Treasuries are headed over the past few days, as my recent two posts attest to.

While I am generally thinking about the benchmark in particular, another way to think about the same question is to look at where the yield curve is headed. CBT bi-weekly expectations survey publishes 6-month T-bill and 5-year bond expectations. The difference between rate expectations of the two instruments 3 and 12 months down the road yields a simple expected yield curve:

Rather than the yield curve, we could simply where markets see rates heading (up or down) in the future by taking the difference between 12 and 3 month expectations for the policy rate as well as 6-month T-bills and 5-year bonds:

Here, you can think of the difference between the policy rate and market rate curves as a very crude measure of markets' expectations of the effectiveness of monetary policy. For example, if markets expect the policy rate to rise by 1% between 3 and 12 months ahead, and market rates only by 50bp, than they expect a less than one to one mapping from policy to market rates.

Note that this is one of the many imperfect substitute indicators I am using in the absence of better data. I can tell you, as someone who used to fill out these surveys, not to make too much of these expectations, especially the currency and rate forecasts. I wouldn't base my investment decisions on these indicators alone, but they are just to have around...

Thursday, August 20, 2009

(un)real rates and sur(real) economists

The ultra-low interest rates are being hailed by many economists these days. Some are even claiming that the policy rate could go as low as 5%.

With the CBT's latest cut, the ex-ante real policy rate now stands at 1.44%, a historic low for Turkey. The low-interest group counters that in many countries, rates are zero, or even negative. The problem is that such naive comparisons ignore country risk. Adjusting for country risk with CDSs reveals that Turkish policy rates do not have much more room to go, maybe another 50bp or so, which is in line with the 75bp cut embedded in expectations. And that is after noting that Turkish CDSs are low compared to the country's credit rating. But this doesn't mean much, as one of the main arguments of the low-interest camp is that Turkey's credit rating is too harsh to begin with, an argument I would agree, if I didn't know the dire state of fiscal balances in the medium-term- but that's a topic for another post.

International comparisons do not help much. While other flexible exchange rate, inflation targeting countries like the Czech Republic and South Africa have 0 real rates, Brazil's, Turkey's closest peer in my humble opinion, is higher than Turkey's. But then again, Brazil does not have an as large output gap.

There is some more room for policy rates, but not that as far as 5% some are recomemnding. That would be as surreal as the economists recommending that path...

Wednesday, August 19, 2009

What is eating bond rates?

The benchmark rallied from just below 10% to 9.72% today, so it might be a valid question to ask how much further can it go. I have four approaches to tackle this question, assuming that the Bank has two more cuts, one 50bp another 25bp, in its sleeve:

1. The difference between rate expectations today and a year from now gives a rough idea on where the benchmark is headed. This approach yields a trough of 9.2%.

2. Equating funding costs to the duration of the benchmark, I end up with 9%, assuming a 100bp difference between the funding cost and the benchmark.

3. Cross Currency Swaps: Now, we calculate policy rate expectations from the CCS market and proceed similar t0 1. This method yields a minimum rate of 9.4%.

4. Integrating all this together, I can also use my ARDL model from last year: It yields a minimum benchmark of 9.2-9.3%.

Sorry for the short explanations, but I am really sleepy. But I would love to discuss these further with anyone interested.

Cyclicality or Circularity?

I have been seeing two really weird arguments in Economics circles recently.

First, it is being argued that Turkey's dismal growth performance is cyclical. To counter, I am offering the following graph:

As you can see, the fall in these indicators (and others) starts as early as summer, 2008. Moreover, I am not sure how you'd explain such dismal growth performance from a country that was supposed to, in theory, be most shielded from the global crisis, as Cevdet Akcay explains, if you would not resort to structural problems and policy mismanagement?

Even more weird is the argument is that the fiscal deterioration is cyclical. Again, a picture is worth more than a thousand words:

You decide....

I am not sure if the owners of these arguments are hoping for a ministerial seat (one is not a Turkish citizen, so he definitely isn't), but the worst part of all this is that such encouraging talk might be tempting the government to complacency, leading to an infinite spiral of circularity.

Render unto Ozel what is Ozel's

Ozel is the last name of Denizbank chief economist Saruhan Ozel. I just found out that he talked about in detail in the long-forgotten NEOs (or UFOs) story of the previous millenium in his latest weekly. It turns out he even gave two references from the Economist that tackled the issue at the time:
Down but far from out; 98,06,30
A goat, a rug, special price? 99, 04, 01
I would have linked the articles, but it requires a premium subscription to the Economist, which I do not have. But I would be grateful if anyone who does could email me the articles or just post them as a comment to this post.

I do not know the guy and do not have access to the article (a friend emailed me the references), so if you speak Turkish and would like to have a look, you need to contact him.

More on unemployment

Mary Stokes has a thoughtful comment on my recent unemployment post. In addition to very precise and concise observations on Turkey's demographics, she poses a couple of really important questions:
What form do you see any potential social unrest taking? How much do you think high levels of unemployment will affect election results?
I definitely do not expect people taking in to the streets in the Latin American or French farmer Jose tradition. However, I do expect an increase in the crime rate at the very least. Unfortunately, I can not back this up or substantiate it, as I do not know any papers that tackle these issues (academics take notice). The problem is that TURKSTAT does not release the raw data and even when they do, they do not release the city codes of those surveyed, so you can not do any cross variation or panel. The rationale they give is that releasing the city codes could lead to dangerous conclusions about unemployment dynamics and therefore lead to social unrest (I am not kidding). A nice circularity, isn't it:):):) If you do not take my word for it, ask Insan Tunali of Koc University, a prominent labor economist, who fell out with them big time because of this issue. Anyway, something may come out on the references I gave on the earlier post, be sure to let me know if it does.

Coming to the second question, Zafer Yukseler of the Central Bank has a paper written in Turkish, where he notes that Economics plays an important role in Turkish elections (he does not do Econometrics a la the Fair model as far as I remember, just casual observation). In fact, you could argue that Economics has been vital in shaping the prospects of AKP, both in their coming to power after the 2001 crisis (of course, there were other factors involved as well) and thier diminished performance in the local elections. That's why I claim that they will open up the coffers to counteract unemployment.

Anyway, if you are thinking about digging these issues further, I would suggest you get in contact with the names I have mentioned above or Seyfettin Gursel of BETAM. The latter follows Turkish labor markets very closely.

More on the August 3 Hurriyet column

Since I just posted my columns for the past three weeks to the blog, I have started to get comments from readers following the articles on the blog (as opposed to the Daily News site). Hakan Tarakci, an old friend and die-hard Besiktas fan, has a question and a comment on the August 3 column, which he entered on Facebook.

First, the question:
is there a website which publishes median house prices and median household incomes in Turkey. I want to see if Turkey is more unaffordable than Australia or whether Oz is the world leader.
And the answer: Yes, Garanti has a useful site on prices. As for incomes, TURKSTAT has household surveys, but they wouldn't give the raw data to you even if you scalp them (as in the latest Tarantino movie), as I will allude to in a later post. Cigdem XXXXX, at a university in DC, has some papers on that survey, if I am not mistaken- do an SSRN search and lemme know if you find something useful.

As for Hakan's comment:
Another thing: In Australia, markets are pricing in a 2% rise in the next 12-18 months. Looks like the discrepancy between investors and central banks is similar all around the world.
Very good point; let's wait and see who gives in. I don't know about Australia, but I am quite sure that Turkey is heading to tears in heaven, with the Bank giving in sometime 2010 big time.

Housecleaning done!

I archived all the missing Hurriyet columns yesterday. Now, my August 17, August 10 and August 3 articles are in the blog. You can reach earlier articles by doing a search with my name at the Daily News archive. Older articles are at Turkish Hurriyet's authors archive.

Now that this is taken care of, I can return to regular blogging...

Tuesday, August 18, 2009

EconNews Roundup

Hurriyet Daily News archive is in the process of being fixed, so I am just reporting the notable items of the last few days.

Daily News has a well-written piece on unemployment, to which your truly commented as well (they actually used my post). It is amazing that everyone else has yet to discover that the most trustable source on labor market issues is BETAM's Seyfettin Gursel.

Turkey may become an energy star, according to IEA Chief Economist Fatih Birol.

Businessmen usually like money spread around, but Turks have enough savvy to know that ends in tears eventually. The article also refers to the ISO semi-annual indices, which I have been arguing for a long time that they should be done monthly. Speaking of ISO, they will be releasing the 2009 second-500 Industrial Firms list next Wednesday, on which I am planning a long analytical piece for the Daily News. So stay tuned:)...

Net Errors and Omissions: A very old story reincarnated

I was talking to a "mature" economist the other day, and he told me, casually, that the current net errors and omissions discussion is not new. A similar discussion was going on in 1997-1998, when the UFOs, or Unidentified Financing Objects, were registering numbers much larger in a relative sense:

In fact, he told me that there was a lively discussion in the papers (news, not academic) between economists on where the discrepancy was coming from and whether it should be ignored. Even the Economist got pulled into the debate, tying the figures to smuggling.

Eventually, just like today, the Central Bank, the guardian of these statistics, decided to take on the issue as well. Their take was to tie the UFOs to shuttle trade from Russia and the NIS; at the time a significant item, especially in Istanbul and the Black Sea coast. They ended up including shuttle trade statistics as well in the BOP. When the NEOs registered a positive number the following year, the discussion was laid to rest, and it stayed dormant after the 2001 crisis, until now, as the UFOs were dwarfed by the ballooning Current and Capital accounts.

Just like more than a decade ago, we have got better statistics as a result of the debate. Now, the CBT tracks resident deposits abroad from the Bank of International Settlements, to which I am proud to say I beat the Bank to- I suggested the methodology in June and did the rough calculations in July. Anyway, while these can only explain part of the errors, as long as there is an improvement, I cannot complain:)...

I give up....

It turns out you can not enter a decent text using the two most popular browsers: In Fire Fox, there is the spellcheck problem. In Internet Explorer, spellcheck works fine, but the line break feature is working even though convert line breaks is checked under settings.

This is really like a joke, so I give up. I am switching back to Fire Fox and will try using Fire Fox's own spellcheck. However, any help with any of these two issues would be most welcome...

Technical Blog Question

Blogspot's spellcheck has not been working for some time now. Basically, it highlights almost all the words in yellow and either offers no suggestions or weird suggestions in a strange language (I actually do not think it is a language at all). Interestingly, this is an issue only when I use Firefox; spellcheck works fine in Internet Explorer, where I am writing this post. BTW, I am running on XP and have the most recent versions of Firefox and Explorer.

The problem actually started just before I left for vacation at the end of July. I was hoping it would miraculously get resolved by itself by the time I got back, but it hasn't! Actually, a quick google search has revealed that other people, also on Firefox, are suffering from the same issue. The only solution offered was clearing cache, which did not work for me.

I know that there are indirect fixes such as using the Firefox spellchecker or spellchecking using Word, but none of these is as convenient as Blogspot's, so any suggestions would be most welcome...

BTW, I just noticed yesterday's unemployment post has a bunch of spelling mistakes because of this problem; apologies for that...

Monday, August 17, 2009

Weekly Hurriyet Column: O consumer, where art thou?

Below is the unedited version of my column for this week. You can read the final version at the Daily News website. First, I have to apologize for yet one more cheesy movie reference, but I just cannot help it. This way, at least you are learning something about my cinema taste as well:)

Coming to more serious matters, one important difference between the Turkish consumer her American counterpart is in the latter's indebtedness, which has important implications for the savings behavior in both countries. But that will have to wait for some time, as I have topics lined for the next three weeks- unemployment, Industrial Enterprises, Central Bank pacification, and The Devil's headscarf (you'll just have to wait for that, no spoilers this time around). At creative spells such as this one, I just wish I was writing at the Daily News not weekly, but well, daily, as the paper's name suggests:)....


Having just returned from two weeks of minimal contact with the outside world, I am relieved to see that the economic landscape has not changed much.

Despite the increasing amount of positive signals, the global economy continues to muddle through a slow recovery path, and regardless of some ups and downs, markets are more or less where I left them. On the domestic front, slow recovery continues to be the name of the game as well notwithstanding the false positive mood created by June industrial production. As for markets, Turkish assets have managed to sustain their momentum, with the lira stable and the benchmark rate hitting single digit territory yet again.

The thorny bush in this idyllic landscape is consumption. The representative consumer, who was showing signs of revival before I left, has vanished into thin air in a matter of two weeks with the release of July data on both sides of the Atlantic. In the US, retail sales fell unexpectedly, whereas Turkish consumption indices confirmed the worrying outlook in the previous week’s CNBC-e confidence indices, registering an 11% monthly decline.

There are two important lessons to be taken from these figures. For one thing, both the Turkish and US data have once again reminded everyone that the effect of temporary incentives seems to be, well, temporary (surprise surprise). In the US, the latest retail sales turnout hints that the cash-for-clunkers program not only may have merely brought future car sales forward, but could also be sucking money from other potential purchases, as sales excluding autos fell a resounding 0.6% month-on-month in July. In Turkey, we have been seeing both effects in last month’s consumption indices.

But more importantly, it should have been crystal clear to all by now that a strong recovery will be impossible without getting the consumer on board. The June Turkish industrial production data is a case in point. A rough sectoral analysis along with the consumption indices of the last couple of months reveals that destocking has run its course after tax incentives have sent consumers bargain-hunting for a while. As for the headline figure, while many were deceived by the improvement in the year-on-year number, adjusting for seasonality and working days reveals a slightly more muted monthly rise than May’s 0.8%, supporting my concerns.

In fact, just merely a day had passed when July capacity utilization cooled down the industrial production euphoria. Taken together, capacity utilization, industrial production and the consumption statistics all point to strong domestic demand headwinds, whereas preliminary exports data for July and August from Turkish Exporters Association hint that global demand for Turkish goods is not recovering at all, as they point to an exports performance actually worse than the first half of the year.

It would be possible to sail against such strong domestic and global headwinds with a comprehensive economic program to steer expectations. But I still see no concrete evidence of that. On the contrary, the latest knockoff fiscal measures and the increasingly worrying medium-term financing outlook mean that the government is intent on going sternpost into the no-sail zone. Therefore, these recent data releases place my forecast for the growth contraction in the second quarter barely in single digit territory, which would translate to a marginal quarter-on-quarter increase, somewhat worse than market expectations.

It will eventually be the consumer, in Turkey or in the US, who will decide the shape and speed of the recovery. The optimist in me wishes to think that she has just taken a temporary respite, but the evidence unfortunately hints otherwise.

Unemployment down for the first time

TURKSTAT just released its labor force statistics for May (April-June period). Let me play the soothsayer, and tell you tomorrow's newspaper headlines: Unemployment down for a third consecutive month. While technically, correct such an approach totally ignores seasonal effects, as I had explained last month. The easiest way to weed out seasonality is to look at the yearly chnage in the unemployment rate, and when I do that, I get the following graph:

Looked in this way, you can see that both the headline figure and non-farm unemployment are down for almost a year.

A slighly more detailed analysis (still requiring nothing more than basic algebra) reveals that out of the almost 1.2 million joining the army of the unemployed, more than two thirds is due to increased labor force participation. While it would be natural to expect the wife and the kids to look for jobs if the breadwinner of the family is out of work or risks losing his job, these dyanmics also reflect the country's demographics. With a young population, Turkey would need a growth on the order of 5% at the very least to stabilize unemployment.

I think this is a challenge that is being ignored by many, for two erroneous reasons. First, it is assumed that rising unemployment is cyclical, meaning that as the country is getting out of recession, the unemployment rate will come down.

To start with, such an argument ignores the main reason why unemployment is a lagging indicator in the first place: There are inherent asymetries in the hiring decisions of companies. Firms will be reluctant to hire unless they know they have strong tailwinds behinds, opting to increase hours worked of existing workers in the meantime. The ongoing lackluster shift from inventory rebuilding to consumption, investment and exports, to which I have been referring to in columns, will slow down this process further. Moreover, when it does finally come down, unemployment could stabilize at permanently higher rates.

Notwithstanding the social unrest associated with a permanently higher plateau, such a shift in unemployment could, in turn, transform it into a leading indicator as consumption is further undermined by more uncertain unemployment prospects. The government will be forced to intervene, especially with general elections looming. And to make matters worse, fiscal policy is likely to be less potent than anticipated in that situation, throwing the government into a deeper maelstrom. And with the state of public finances, this will all start looking rather ugly, as unemployment and growing public debt usually do not go well together.

I am getting ahead of myself (and the data). After all, today's numbers are marginally better than previous couple of months. But as the elders say, it usually pays to be overly cautious now rather than sorry later.

Sunday, August 16, 2009

I am back...

from two weeks of sailing in the Southern Aegean. In the next couple of hours, I'll be archiving my Hurriyet columns for the last couple of weeks. Then, I have a couple of interesting things I have been dying to write about.

Monday, August 10, 2009

Weekly Hurriyet Column: For tourism the bells toll

Below is the unedited version of my column for this week. You can read the final version at the Daily News website. FYI, there was an interesting article in the Economist that recounts Europe's tourism woes. While they are right on an absolute sense that the European tourism destinations have been hurt real bad by the crisis, Turkey has actually fared worse on a relative sense, as I argue below.

The article is the first time a graph from my friends at Turkey Data Monitor, an integral part of my blog, has been featured in Hurriyet. I do not want to spill all the beans, but you'll soon be seeing a lot of their stuff in the Daily News, a partnership that has been on the agenda for some time, but one which has been delayed due to summer vacation plans and the paper's efforts to improve the web site. So, as I always say, stay tuned...



As I will be sailing in the Aegean this week, I thought it’d be appropriate to devote the weekly column to the tourism sector.

But such a title all but ensures that you will think I have had a sunstroke. After all, the media reaction to the latest tourism statistics was almost unequivocally positive. With the world economy still at a slump, the 1.3% yearly decline in tourist arrivals for June could easily be regarded as a resounding achievement. Substitution from countries with swine flu and the weak lira were touted as the main reasons behind the strong figures.

While those factors might have played a role, a closer look does not paint such a rosy picture. For one thing, the latest figure compares less favorably to the same months in the previous two years. When compared to 2007 and 2008 June yearly figures of 17.1% and 19.2%, the 2009 turnout looks rather dismal.

Likewise, a cross-country look at the numbers reveals, at first, that Turkish tourism has been rather resilient compared to Mediterranean peers like Greece, Spain and Croatia, most of whom are seeing declines in double-digit territory. However, while the falls in these countries have been larger in an absolute sense, they have nevertheless not seen as large departures from their recent trends. In fact, the 20.5% decline from last June’s growth rate is second only to Croatia’s.

Looking at year-to-date data and world trends reveals an analogous picture. The 0.9% yearly decline in arrivals in the first half of the year is slightly better than World Tourism Organization’s forecast of a 1-2% fall in global tourist arrivals for 2009. This is really subpar performance for a country that has been increasing its share of the global pie in the last decade. To form an analogy, just think of China growing 5% in 2009; that would be perceived as devastating news, but a much larger relative decline in Turkish tourism is packaged as a success.

When you look at revenues, more disturbing signs emerge. Despite seasonal differences, revenues per tourist have been on a downward trend since 2003. As a result, the robust arrivals figures have not been reflecting as strongly into revenues for the last few years. For example, June revenues were down 8.7%, with most of the decline coming from the fall in per tourist expenditures. Anecdotal evidence puts the blame on lower prices and shorter stays, but I doubt this would be enough to explain the emerging pattern of the last few years. For now, this is an enigma that begs to be explored further.

There is one part of the tourism statistics giving me a glimmer of hope: The outgoing tourism data. While evidence is again anecdotal at best, the number of Turks going aboard, after falling as much as a half (on a yearly basis) early in the year, is back at last year’s levels. This is yet another piece of evidence that the economy is continuing with its ascent after hitting rock bottom in the first quarter.

I guess the title was a bit too strong after all. The tourism sector has not been shattered by the global crisis. But the worrying long-term trends definitely warrant attention.

Saturday, August 8, 2009

On last week's Hurriyet column

It turns out I made quite a good call with my bond outlook last week, but that wasn't really rocket science after the Inflation Report. I am on vacation and too lazy to write my own thoughts, so I am just copy pasting a well-written FT article on this week's incredible rally. BTW, if you have been following my columns, you already know where I stand, but from those interviewed, I am more or less close to Murat Ucer's view. Anyway:

Bond boost for Turkish Treasury By Delphine Strauss in Ankara Published: August 5 2009 17:49 | Last updated: August 5 2009 17:49Yields on Turkey's sovereign bonds fell to an historic low this week as investors took advantage of extreme dovishness from its central bank and a return of risk appetite in emerging markets.The rally helped the Turkish Treasury sail through some of the heftiest debt refinancings it faced this year.But economists warn that central bank policy will not be sustainable unless Ankara rapidly commits to credible fiscal retrenchment.The Treasury successfully borrowed TL10.8bn ($7.4bn) in non-competitive sales and auctions of 21-month and five-year bonds.It had already placed Treasury-bills totalling TL5bn on Monday at a compound rate of 9.54 per cent.Yields on the benchmark 21-month bond dipped below 10 per cent for the first time ever after the auctions, which have taken the Treasury much of the way towards its target of borrowing TL17.3bn in August, against redemptions of TL17.4bn.Yields on the JPMorgan Government Bond Index-Emerging Market Turkey reveal a dramatic drop from 24.6 per cent last October to 10.4 per cent this week.The smooth sales mean government borrowing "has become a non-issue for financial markets", said Ozgur Altug, analyst at BCG Partners in Istanbul, who noted that the Treasury had hoarded foreign exchange reserves ahead of the redemption period.The exuberance extends to equity markets.Istanbul's main share index hit a fresh 17-month high this week after an extended rally erased not only the losses sparked by the global crisis but also some of those sustained in political instability earlier last year.But bond investors, above all, are responding to unequivocally dovish signals from Turkey's central bank, which has already cut interest rates by 850 basis points since November.The bank took markets by surprise last week with a quarterly inflation report suggesting it expected to cut rates further and then hold them throughout 2010.Christian Keller, economist at Barclays Capital, said this week's data showing inflation had eased as low as 5.4 per cent could encourage the bank to cut its benchmark borrowing rate a further 100bp or more in the coming months.That will take monetary policy into uncharted territory in a country with a history of repeated economic crises, double-digit inflation and currency collapse that forced the central bank to keep interest rates among the world's highest to stem capital outflows."If [the central bank ] has help from the government, they'll come out of this crisis as a developed market . . . They want to reduce the differential between Turkey and other markets," Mr Altug said.But for its policy to be sustainable, Turkey's government will have to deliver on promises to restore fiscal discipline, after a year of rapidly widening budget deficits, say many economists.Officials are working on what some commentators consider to be more realistic medium-term fiscal plans, but many doubt they will be enforced unless Turkey accepts International Monetary Fund policing.Some think the central bank is taking risks in pre-empting any commitment to long-delayed fiscal reforms."I think it's sustainable for another month or so. Global risk appetite is back, and no one has the luxury of shunning Turkey for problems that may emerge in November and December. But I find the central bank's approach very puzzling," said Murat Ucer, Istanbul-based analyst for the consultancy Global Source."It is clear that this extreme easing will have to be removed at a point in the not so distant future," said Ahmet Akarli at Goldman Sachs, noting that most "healthy" emerging market central banks were becoming nervous about exit strategies.Erhan Aslanoglu, economist at Marmara university in Istanbul, also says the central bank may be too optimistic in its assessment of how long disinflation will continue.But he says it may still achieve a lasting reduction in Turkey's historically punishing domestic borrowing costs."There may be a new equilibrium – not as low as it's going to be by the end of this year, but not as high as before the crisis."
Sent by BlackBerry Internet Service from Turkcell

Monday, August 3, 2009

Weekly Hurriyet Column: Something’s gotta give

Below is the unedited version of my column for this week. You can read the final version at the Daily News website. As I noted from the comfort of a sailboat a few days after the column was published, it is the markets that are giving in at the moment. I hope this lasts...


The Central Bank (CBT) shattered my resolve to devote two weeks to long-term structural issues in tourism and industry with its latest Inflation Report.

While a continuation of the Bank’s dovish stance was the consensus view even before the summary of the Monetary Policy Council (MPC) minutes was released, last week’s report is as dovish as it gets. The Bank’s base scenario sees policy rates steady until the end of next year after some more easing. In fact, with the implementation of a medium-term sustainable fiscal framework, rates could be in single-digit territory for up to three years.

The Bank’s economic rationale is clear: Both the world and Turkey are to recover very slowly. Not only sustained low global rates are one of its key assumptions, the CBT has also revised down its domestic growth forecasts. However, the Bank (and many economists) sees a strong recovery in the output gap for 2010, which could put some pressure on inflation and consequently on policy rates. Nevertheless, there are other reasons for the Bank’s stance.

A former MPC member used to joke that his CBT ID had a Treasury Personnel stamp on it. With almost two thirds of the August Treasury redemptions of TRY 25 billion coming this week, this gag has taken on an entirely new meaning. With a seasonally high primary surplus and its coffers full, the Treasury’s hand would have been strong even without CBT support. But the planned rollover ratio of slightly less than 100% could be exceeded at this week’s auctions with very strong demand and record-low rates, allowing the Bank to build its war chest further.

The sustainability of these low rates or whether they will be able to affect the real economy are entirely different matters. On the former, the Bank has made it clear that market rates are out of sync with its policy outlook. Despite the strong rallies after the Inflation Report, both Treasury yields and cross currency swaps are still pricing in higher rates a year on and for end-2009. With survey expectations having penciled in a full percentage point rate hike in the next 12 months before the Inflation Report, this week’s survey has suddenly come to the spotlight.

The CBT is not the only Central Bank faced with this problem. Despite Bernanke’s latest assurances, markets are pricing in rate hikes of 1.5% until the end of next year, and long-term rates are still very high. This has led the chairman to a roadshow, as he has been appearing at anything from 60 Minutes to a Kansas City town hall. His failure would violate one of the key assumptions of the CBT. You can see the CBT Inflation Report as a similar effort to talk rates down.

The CBT’s job is arguably tougher than the Fed, as markets are not buying into its long-run inflation outlook, either. While both the Bank and markets see end-year inflation at around 6%, their projections move in opposite directions after that. This week’s expectations survey will be important in that respect as well.

But my real worry is not expectations. There is now a serious risk that these measures could backfire. Not only the Bank is running, without a credible fiscal framework, the risk that its stance will be perceived as fiscal accommodation or debt monetization, it could also lead to asset price bubbles a la Turca, with the banking sector and bonds being the obvious candidates. Dollarization would be another consequence, which would ease the pressure off the lira a bit, but at the expense of some dear liquidity.

At the end of the day, with so much out of sync, either the CBT or the markets will have to give in. In case it is the Bank who ends up caving in, I have two suitcases packed so that I can close shop and leave for an exotic island right away.

Sunday, August 2, 2009

Me, myself and I (and the blog) on vacation

I will be on vacation for the next two weeks. While I will have my Crackberry on me and Bberrying to the blog from time to time from the comfort of a sailboat in the Aegean, I should have only sporadic entries.

I will be back in full force in the third week of August.

Saturday, August 1, 2009

EconNews Roundup

Yet more IMF mambo jumbo- this time from the business perspective

This is all from Hurriyet, as the Economics section has no news from Tuesday to Thursday for some weird reason. Unfortunately, those were the hottest days in terms of Economics. I was planning to include a couple of items on the MPC minutes, Inflation Report, Trade Statistics and the like, but there is absolutely nothing there on that. Therefore, as painful as it may be, I have to turn to the competition:

Zaman has a very engaging interview with Yapi Kredi Chief economist Cevdet Akcay. I had summarized his controversial (actually, they are controversial because the absurd has become the norm, the illogical the haute couture in terms of economic thinking in Turkey) views on the Turkish savings rate a month ago. Read the whole thing, but I found most interesting his description of a crisis-prone country, which accidentally fits Turkey really well. In other words, the PM was half right after all: The crisis was supposed to pass Turkey tangent on a theoretical basis, but it ended up as a diameter because of bad policy management.