The National Bunk of Kazakhstan

Is the central bank truly targeting inflation?

Just when you thought it was safe to go back into the water….

On Monday, Kazakhstan’s central bank cut the base interest rate by a hefty two percentage points, to 13% per year. The National Bank’s statement justifying this decision is curious. It does not dwell on the reason for the cut; presumably the Bank wants to spur demand for the nation’s products by making loans more affordable. In the first half of 2016, production was down .7 of 1% when compared to the same period of 2015, according to the government’s statistical committee.  Particularly sharp was the drop in exports (minus imports) – almost 5%. The Bank faced political pressure: Before it acted, the minister of the national economy, Kuandik Byshymbaev, had said he favored a rate cut because inflation would die down by the end of the year. Some analysts also maintain that Britain’s surprising vote last month to leave the European Union has created anxiety among investors who are best consoled with cheap money. Whatever. It’s a small planet.

The Bank devotes most of its statement to explaining why the rate cut won’t steep the fires of inflation. At present, consumer prices in Kazakhstan are surging by more than 17% per year.

The Bank writes, “Inflation is fully corresponds to the forecasts of the National Bank, and the risks of acceleration of the inflation under current conditions are minimal, which with a high probability allows expecting inflation to achieve the upper limit of the 6-8% inflation target band by the end of 2016. Taking into account the time lag effect of the base rate on inflation, which is estimated to take up to one year, the decision to reduce the base rate confirms the confidence that inflation will remain within the target band over the horizon of twelve months and up until the end of 2017.”

No, I didn’t doctor that quote. I took it word for word from the Bank’s press release Monday. Central Asia’s leading central bank believes that neither good English nor clarity nor logic matters. Nevertheless, let’s see what we can do with what we have to work with.

Last December, consumer prices were 13.6% higher than in December 2014 (the Bank refers to this as a “year-to-year” change).  In every month since then, the year-to-year change has only increased; it was 17.3% last month (Table 1). Usually, “acceleration” means an increase in the rate of change of the rate of change, although the Bank’s English tends to be rather creative. In June, the rate of acceleration in inflation was .2, the highest so far for 2016. Despite the Bank’s claim, the “risks of acceleration of the inflation under current conditions” are not “minimal.” Inflation is accelerating right now.

Month	Inflation rate (%)	Rate of change in inflation	Acceleration of inflation
December 2015	13.6
January 2016	14.4	.8
February 2016	15.2	.8	0
March 2016	15.7	.5	-.3
April 2016	16.3	.6	.1
May 2016	16.7	.4	-.2
June 2016	17.3	.6	.2

Table 1: Inflation in Kazakhstan

Source of data: The National Bank of Kazakhstan

Let’s move on. The Bank says inflation will probably return to the target band of 6-8% by the end of the year – and so there’s no harm in spending a bit more. But the Bank keeps moving the goalposts. In October, it said it would bring inflation back within the target band “in the medium term.” Well, the medium term is here and the inflation rate isn’t there – not, at least, the year-to-year rate, which has the advantage of adjusting for seasonality by comparing rates of the same season. In May, the Bank said “the forecast of annual inflation for 2016 remains within the band of 6-8%.” Hello? The year-to-year rate has been nowhere near the band since September, and it’s moving further and further away.

Torturing the data until it confesses

It’s hard to know just what the Bank is up to. Suppose that we adopt the year-to-year definition of inflation. Then a generous interpretation of the Bank’s vows is that it will deliver a December 2016 price level that is no more than 8% higher than the December 2015 level.  (The “price level” is the price of a typical bundle of goods and services.  Think of it as the “average price” in the economy.) This June, the level of consumer prices was already 4.6% higher than the December level, according to data reported by the Bank (not adjusting for seasonality; see the Notes). If this trend continues, the Bank will miss its target again, but not by much. The price level in December 2016 would be 9.4% higher than in December 2015.

More troubling is the trend in all prices, not just those paid by consumers.  The GDP deflator indexes prices comprehensively, including the input prices paid by producers. For the first quarter of this year, the deflator indicated that prices were 12.7% higher than in the first quarter of 2015, according to the statistical committee.  This matters: Inflation may show up in input prices first and in consumer prices later, since producers set output prices that will compensate for increases in wages and capital rents. Judging from the deflator, and from the fact that consumer prices are rising more rapidly than earlier this year, the Bank is not likely to hit its target.  

NBK is grasping at straws. In its press releases, it does contortions to show how it is satisfying its inflation targets, even though prices are actually galloping into the sunset, and how things will be fine in a few months, just wait and see. If the Bank truly wants transparency, it should admit that its inflation policy just might be toast. –Leon Taylor, tayloralmaty@gmail.com

Notes

To calculate inflation, I have used the year-to-year change in the price level. For example, if the price level is 100 in July 2015 and 104 in July 2016, then the annual inflation rate is 4%.

Especially in press releases, the National Bank often uses computations of inflation that accumulate monthly changes in the price level. For instance, the monthly change in consumer prices was .3 in May 2015 and -.1 in June 2015. Then the rate of inflation from (the beginning of) May 2015 to (the beginning of) July 2015 can be derived from (1.003)*(.999)*100, where 100 is the base value of the price level.  The calculation gives us (1.003)*(.999)*100 = 100.1997. That is, the price level at the beginning of July 2015 was about two-tenths of 1% higher than at the beginning of May 2015.

Extrapolating this two-month result to 12 months gives us an estimate of the annual rate of inflation (said to be “annualized”). Let’s do the numbers: (1.001997)^(6)*100 = 101.2042.  The annualized rate of inflation is 1.2%. In general, to annualize an n-month estimate, compute (1+x)^(12/n)*100, where x is the n-month rate of change (in our example, .001997).

This algorithm is prone to several problems. First, it computes only a short-run change in prices (over just two months), which may not characterize the actual long-run change (over 12 months). If oil prices suddenly rose over the two-month period but leveled off over the next 10 months, then annualizing the two-month rate of inflation may overstate actual inflation.

Also, month-to-month changes are small, so a rounding error of a given size may produce a relatively larger error in the monthly calculations than would have occurred in yearly calculations.

Finally, monthly calculations may depend on the season. In our example, one month (May) occurs in the spring, and one month (June) in the summer. Spring pricing may differ from summer pricing for various reasons.  People vacation in the summer, which may lower domestic demand if they travel out of the country and are not replaced by in-coming tourists. Teens still in school may look for summer jobs by offering to work for less pay, which can reduce the price level. And so forth.

Whether to adjust price-level estimates for seasonality is controversial, and it depends on the purpose of the estimates. If we want the long-run trend in inflation, adjusting for the season will give us the simple trend.  But if we want the actual rate of inflation, then we shouldn’t adjust for the season, since the fact is that seasons affect prices.

In any event, one might largely avoid the problem of seasonality by comparing the current price level to the level that occurred in the same month last year.       

References

Nikolai Drozd. Ключевой понедельник. [Key Monday.] Курсивь. July 8, 2016.

Kazakhstan statistical committee.  National accounts – integrated accounts.  2016. www.stat.gov.kz

Kazakhstan statistical committee.  Prices and tariffs.  2016. www.stat.gov.kz

Mankiw, N. Gregory.  Principles of economics.  Third edition.  2004.  Includes a good discussion comparing the consumer price index to the GDP deflator.

National Bank of Kazakhstan.  Price index.  2016. www.nationalbank.kz

National Bank of Kazakhstan.  Various press releases on the base rate. 2016.  www.nationalbank.kz