It Depends

One of the current success stories of late for Rhode Island has to be retail sales. Not only has retail sales been in a well-defined uptrend since the last quarter of 2009, at the present time it is very close to its highest level since 1990.

Retail sales can be either a coincident or lagging indicator, meaning that it either reflects what is going on at the present time or what has recently occurred. For Rhode Island, the most timely data on retail sales is that for retail sales subject to the sales tax. Based on when these taxes are paid and recorded here versus when the sales actually took place, retail sales for Rhode Island is a lagging indicator. Also, retail sales is a very cyclically sensitive indicator, which means the value of retail sales is very sensitive to the overall level of economic activity. The technical term for this is that retail sales are pro-cyclical, since they move in the same direction as the economy.

Like any economic indicator, it too has its limitations: since this is based on retail sales tax collections in Rhode Island, it excludes any items that are not currently subject to taxation, most notably clothing. And there can be distortions. For example, given the very mild winter, winter clothing expenditure very likely fell since the same time last year. If that decline was sufficient to move total retail sales lower than they were last year, the understatement from excluding clothing sales from our sales tax data might not be as large as it typically is. Similarly, since the meals tax in Rhode Island has been 14.7% higher than the regular sales tax rate for several years now, this has distorted "retail sales" for meals over this period -- tending to overstate the overall total. While this may sound like minutia, it is important to understand that every economic indicator has its share of "baggage." I have believed for quite some time now that the indicator any person likes best is probably the one that person knows the least about.

Let's take a look at the historical performance of retail sales in Rhode Island, using as our basis quarterly data on retail sales subject to the sales tax expressed at an annual rate. The chart below (click to enlarge) shows values since the first quarter of 1990. Please note that the vertical axis is stated in logarithms, mainly because a given vertical distance represents the same percentage (relative) change between any two period in the chart.

The most recent peak in retail sales occurred in the fourth quarter of 2006, at $12.7 billion. Can you think of anything else that happened that quarter (hint: we have discussed this numerous times in prior posts)? The answer: it coincides with Rhode Island's most recent employment peak. This reflects the fact that retail sales are very sensitive to the overall level of economic activity, as stated earlier. The trough for retail sales was in the fourth quarter of 2009, at $11.2 billion. Thus, Rhode Island's peak-to-trough decline in retail sales was 11.8 percent. Note that retail sales have trended upward throughout this recovery, although not straight up. Once again, the cyclical nature of retail sales is in evidence here.

This chart provides an excellent illustration of the mindset of most, if not all, of Rhode Island's leaders, its media, and its residents: things here have now turned around nicely, and Rhode Island is doing much better it has in a very long time. Add to this the fact that income tax revenue is up, at least above expectations, and if you read my last post, employment is apparently higher and our state's jobless rate is lower than the currently published data indicate. Thus everything appears to have returned to normal.

I agree totally with the assessment that everything here has returned to normal -- but Rhode Island's "new normal." In economics, we never look at things in absolutes. Instead, we always evaluate things in relative terms. This precludes our making substantive judgments about the behavior of indicators that are stated in current dollars, as retail sales is above. So, let's now turn to evaluating retail sales here in relative terms -- meaning relative to the purchasing power of the present time. This, recall, means examining the historical behavior of real retail sales.

The next chart shows the behavior of real retail sales subject to the sales tax for Rhode Island since 1990 (click to enlarge). As should become readily apparent almost immediately, this gives a very different picture of our current situation.

When taking inflation and purchasing power into account, the peak in real retail sales occurred in the third quarter of 2004, at $14.8 billion. This peak thus came well before the current dollar peak (in 2006Q4). Note that by stating retail sales in the purchasing power of Q1 in 2012, we are "inflating" current dollar values from earlier periods into what they would be if dollars then had the same purchasing power as today, which explains why earlier current dollar values are higher when stated in real terms. The trough in real retail sales occurred in the fourth quarter of 2009, at $11.9 billion. This is the same quarter as the trough in current dollar retail sales.

The reality of Rhode Island's current situation, our "new normal," to quote the phrase coined by Mohammed El-Erian of PIMCO, is in fact very different from the predominant view of our state's economic performance. First, real retail sales here have been falling since the third quarter of 2004, well before our most recent employment peak. Second, the peak-to-trough decline in real retail sales here was much greater than one would conclude based on current dollar values, just under 20 percent. Third, while real retail sales have improved since their trough in 2009Q4, they have registered almost no gain since that time: $12.1 billion versus $11.9 billion. Finally, the most recent trend in real retail sales is sideways to downward. Thus, Rhode Island's economy has failed to perform well enough to consistently generate retail sales growth rates in excess of inflation for quite some time now.

Where does all of this leave us? Not where almost everyone in this state appears to believe. Retail sales data, when viewed in the correct context of real values, confirms something I have written about in more posts than I care to think about, that Rhode Island is in a very tepid recovery. So, even when (or if) current dollar retail sales here exceeds it recent high of $12.1 billion, we will not have returned to anywhere even remotely close to where we once were. So, do we continue to contrast the current situation with the depths of where we once were, as seems to be the dominant view, or should we shift our focus to where we are relative to recent heights? You can guess which I will continue to opt for.

We should be attempting to move toward recent highs -- that should be the standard of excellence or satisfactory performance here. And, guess what? That outcome doesn't just fall on us from heaven. We have to work for it. And very often, this will require that we work very hard to get there. I heard a great saying a few months ago that pertains to this: "Success is not a destination, it is a journey." It's time for Rhode Island to purchase its ticket for this journey, because the dominant perspective here, to dust off an old phrase I used to use to describe Rhode Island about a decade ago, defines success in terms of "Relative Mediocrity."

Retail Sales in Rhode Island Since the Last Recession

One of the most critical elements that determines Rhode Island's economic momentum is retail sales. Retail sales drives a great deal of economic activity here while itself being determined by how well the overall economy does. Remember, too, that sales tax revenue is an important source of Rhode Island's overall tax revenue.

Retail sales have done better since Rhode Island emerged from its last recession in February of 2010. But, like so many other economic measures, its behavior during this recovery has been choppy at best, reflective of the fact that our state's rate of growth has recently begun to slow.

The best way to analyze retail sales is to take inflation into account. Doing this we obtain real (i.e., inflation-adjusted) retail sales. I have calculated these so that the most recent month of data, April 2011, is the base period (the basis for comparison). The chart below shows real retail sales for Rhode Island since 2007 (click to enlarge).

The most striking feature of the performance of retail sales since 2007 is how far they fell. A quick look at the graph also shows rather disturbing trends not only in their rate of decline but the duration over which their decline occurred. All of this is a testament to how severe the last recession was. It wasn't given the name "The Great Recession" for nothing! From this chart, it should also be fairly easy to grasp one of the reasons why tax revenue here declined during the last recession and has only recovered a bit during this recovery.

During the last recession, real retail sales moved into a downtrend in June of 2007 that lasted for over three years. The "breakout" from this prolonged downtrend, which coincided exactly with the beginning of Rhode Island's current recovery, occurred in February of 2010. 

In more "normal" times (I have forgotten what those actually are by this point), we would have seen a fairly rapid rebound from such a precipitous decline in retail sales. But in those times "leverage" was rampant as credit, whether credit cards or home equity lines of credit, were used without a second thought, banks were very generous in their lending, mediocre credit scores weren't much of an impediment, and collateral was something that persons had to have back in the dark ages. But that was then. The chart above shows now -- not a very substantial recovery in real retail sales. After an initial jump, real retail sales began to fluctuate, largely sideways but trending slightly downwards, meaning that the value of retail sales has barely kept up with inflation. The same has been true for manufacturing wages here (see previous blog post on this). The recent slight downtrend also illustrates that the pace of economic activity in Rhode Island has been slowing of late.

Will real retail sales here break out from the most recent downtrend? Several factors could allow this to occur, namely low interest rates and ongoing national and state recoveries. However, the low and declining interest rates themselves reflect a slowing in the pace of national economic activity. Add budget problems in RI (sorry to be redundant) and for the US, and pension woes that must be addressed, and it becomes apparent that it is largely the underlying cyclical momentum of the US and Rhode Island economies that will largely determine whether a breakout occurs. Should these factors fail, our state faces the prospect of a breakdown below the recent trend, potentially eliminating some or much of the relatively small recovery gains we have experienced over the last year. Should that occur, revenue "surprises" will quickly become disappointments. Let's keep our fingers crossed that this is not what occurs. We already have enough on our state's plate as it is!