Measuring savings

Today's Wall Street Journal online has a report on Is the Negative Savings Rate A Negative for the Economy? which includes this graph:

The online debate is a lively exchange as to whether the negative savings rate will be a long term detriment to the economy. There is only one problem with this debate, as one of the participants pointed out: we may not be measuring the savings rate correctly. Savings is not measured directly, it is a residual of what is left over when we account for income and spending. It is subject to assumptions on what is spending and what is investment. And, in fact, BEA estimates that "if R&D were included in the GDP as investment instead of as an expense, business investment would be 11 percent, or $178 billion, higher; and the 2002 national savings rate would be 16 percent instead of 14 percent." (See my earlier posting on this.)

That is not to say that the overall direction of declining trend shown above isn't correct. Nor does it mean we should be concerned. It does mean that we don't know for sure. In the period of beginning in the mid 1980's -- when the trend line heads south -- the US economy what being transformed from an industrial to a knowledge economy. That change means our investment patterns have also changed. More and more of our investments are in knowledge and intangibles, rather than plant and equipment. But our account rules have not changed. Plant and equipment gets depreciated over time as an investment; knowledge creation gets expensed immediately. As a result, our savings rate is distorted (since investment gets counts as part of savings, not spending).

So, the question remains: How much of that decline in the savings rate is real and how much is an accounting issue? Before we starting reading too much into the numbers, let's figure out whether they are, in fact, correct.