Bond yield comparisons are seriously distorted by using consumer price index deflators.

Brian Sturgess - February 2013

Speed Read
  • Measuring real bond yields using GDP deflators shows a radically different picture from using consumer price inflation indexes, the use of which can seriously distort investment decisions
  • The GDP deflator measures general inflation across all domestically produced goods and services and not a basket of consumer goods
  • The highest real (GDP deflated) yield recorded in the top economies was for France (4.6%). Using the “normal” CPI deflator the French government bonds yields were 0.4%.

Real interest rates are simply the difference between nominal yields and inflation, but the price index used to measure inflation can have a significant distorting impact on investment decisions. The most appropriate price index to use is the GDP deflator, but the most frequently used in practice are consumer price indexes. In March 2012, a report by PricewaterhouseCoopers to the UK’s Financial Services Authority recommended using the GDP deflator instead of the Retail Price Index for projected rates of return on a variety of asset classes.[1] On the basis of lower economic growth projections and the change in deflator the FSA subsequently significantly downgraded its projected real rates of return due to become mandatory in April 2014[2].

Apart from the impact on the financial services industry the distortions arising from the wrong choice of index can seriously impact upon differences between real government borrowing costs and the burden on taxpayers. Even small differences between the CPI and the GDP deflator can have a large impact on the real rate that a government is borrowing at compared with the often consumer price indexed linked welfare payments it makes.

The impetus to use consumer price indexes arises from governments who build them into index-linked government securities which informs much market comment on the comparative real yield on other bonds. In the US, for example, Treasury Inflation Protected Bonds (TIPS) pay interest rates that are adjusted using the Consumer Price Index (CPI) while in the UK index-linked gilts are based on changes in the Retail Price Index, a measure of inflation based on a basket of consumer purchases rather than an estimate of inflationary pressures across the economy.

There are a number of key differences between the scope of the CPI and the GDP deflator and in the methods by which they are calculated. The GDP deflator measures inflation in the prices of all goods and services produced domestically by the private and the public sectors. The CPI measures changes in the cost of purchasing a fixed basket of consumer goods which includes both domestically produced goods and services and imported products. The wider coverage of the GDP deflator makes it more appropriate for use in calculating real government bond yields because it allows a comparison of the real cost of finance with the level of government expenditure in constant prices as a proportion of GDP. In addition, it measures the real return to bond holders in terms of real national income generated and not the real cost of purchasing a sample of consumer goods.

The differences in calculating real government 10 year bond yields with the GDP deflator rather than the CPI is illustrated by data for the ten largest economies in Figure 1. In France, real government bond yields adjusted for general inflation came first in providing high real returns. Ten year French Government bonds provided domestic holders with a real return of 4.6% rather than the far smaller 0.4% suggested by the use of the consumer price deflator. This is far better than the real returns of 0.1% and 0.4% in Germany and the USA respectively, while domestic holders of UK government debt experienced a poor negative real loss of -1.1%.

In practice, the difference between the deflator and CPI for some countries is often relatively small, but as the table shows for some countries using 2012 data to deflate current bond yields it was enough to turn a negative yield to positive – (Germany and India) or to make a positive yield negative (Russia).

The CPI remains the most common price index used in calculating real bond returns, but it is highly recommended that investors insist on their advisors using the GDP deflator for calculating historic and projected real returns.  As noted earlier this was one of the little noticed technical recommendations of a report by PricewaterhouseCoopers to the UK’s Financial Services Authority in April 2012 for projected rates of return.