The NEW LEIPLINE is Here!
What is LEIPLINE? LEIPLINE is a FREE subscription only, quarterly newsletter that examines the details of the economic data that we collect for the Jacksonville MSA to provide greater insight for our readers. Each February, May, August, and November the LEIPLINE will provide an assessment of the just past quarter and will forecast for the next quarter using the data we collect and other pertinent local, state, and national data. The first LEIPLINE will analyze all of 2006 instead of just the fourth quarter of 2006.
In addition, each quarter we will provide research articles, developed by the LEIP students, to assess some interesting implications that the time series of data going back to January 2002 provide. In this first LEIPLINE, beyond the article on 2006, we provide an assessment regarding the impact of the 2005 Super Bowl on local retail sales; and another article regarding the Phillip’s Curve relationship between inflation and unemployment in the Jacksonville MSA.
Each month we publish new information about inflation, leading indicator, stock price, and unemployment data to see how the climate is changing. Then, revisit the LEIPLINE each quarter to gain further insight into the Jacksonville economy provided by our excellent staff of student economists.
Ordinarily, the lead article in each quarterly LEIPLINE will be a review of the previous quarter with a look ahead at the subsequent quarter. Since this is the inaugural LEIPLINE, we look back at all of 2006 and forward for the first quarter of 2007. Our focus each quarter will be on both the four variables for which we collect data and the implications of those data for the Jacksonville MSA overall. Since the local CPI is the most significant variable that we develop we will start with that.
Thr 2006 Jacksonville CPI
The annual inflation rate for the Jacksonville MSA for 2006 was the mildest of all of the years since we began collecting the data in December 2001. The aggregate increase in prices was a paltry one-third of one percent for the year. This is directly comparable to the 2.4% increase for the nation as a whole. The LEIP CPI has consistently lagged national inflation growth suggesting that we simply do not see the same increases in prices in the local area. There are many possible justifications for this realization (including a downward bias in our numbers due to the small relative sample size), but while average prices have grown nearly 14% since LEIP started, they have only grown 4.5% locally for the same time period.
For the year (2006), some very significant changes arose, but they were very different than those that dominated 2005. Household insurance rates rose 29.1%; fuel was up over 28.4%; water, sewer, and trash rates were up 29%; and personal care product prices rose 34.4%; relative to December 2005. Food prices were up an average of 7% (led by cereal and bakery at 11%) with motor vehicle insurance up almost 14%. Down were other apparel (almost 21%), motor vehicle parts and equipment and fees (17.7% and 5.2% respectively) and video and audio falling more than 14.7%. The other categories changed very little for the year. Interpreters of inflation data need to be cognizant that even though many categories revealed significant price increases, when other categories move in the opposite direction, the overall impact may be quite small. Consumers and businesses are always more cognizant of price increases than decrease, so there is a tendency to expect higher rates of inflation than actually occur.
The outlook for the first quarter of 2007 suggests continued moderate price inflation, if not slight declines in prices. The weather has been mild without a significant freeze suggesting that food prices should be stable. Fuel costs may rise with the OPEC production cuts but mild weather and declining housing values should offset any increase. Retail sales tend to jump in January which may stimulate price increases, but this too may be mitigated by continued outsourcing and declining costs in manufacturing and increases in labor productivity in general. We will be surprised if locally prices grow even one half of one percent in the first quarter.
Unemployment rates in Jacksonville during 2006 hit lows that have been rare in the last three decades. April’ rate that was 2.85% is the lowest the unemployment number has been since 1998-99. LEIP only seasonally adjusts the unemployment rate (we do not calculate it), but doing so provides valuable short term insight into the changes from month to month. For example, the unadjusted unemployment rate between December 2005 and January 2006 went up from 3.0% to 3.2% while the seasonally adjusted value fell from 3.2% to 3.02%. January 2006 actually started an extended period of very low unemployment, even though without seasonal adjustment it appeared to be starting a trend back up. The bottom line on unemployment is that job growth in the Jacksonville MSA has been very strong for more than a year, but the influx of new workers has been almost as strong, keeping unemployment rate much lower than the nation as a whole. This outcome is largely due to the service orientation of our local economy and the relative stability of employment in service sectors. We see no reason to believe that the first quarter of 2007 will generate much change in unemployment in either direction, although the housing market slump will be the culprit if unemployment expands much in January through March.
The LEIP leading indicator was up over two points in 2006 and is over 6.5 points higher than the comparable national LEI on which it is based. In particular, the last four months of the LEI were up sharply (2.4 points) suggesting that the second and third quarters of 2007 should be positive. The outlook for the first quarter of 2007 is more uncertain because the LEI was down 1.37 points in the July through September quarter of 2006. However, with the concurrent indicators mentioned above being so strong, we do not anticipate much of a slowdown in the first quarter if at all. Building permits, consumer confidence, and the real M2 money supply will be the keys to growth in both the first quarter of 2007 and beyond. Thus, once again, the housing market will be a driver. In next quarter’s LEIPLINE we will look at the relationship between the LEI and other local indicators to determine how effectively it predicts and with what lag structure.
Stock Price Index
The local stock index had a rough year in 2006. Winn Dixie’s problems were just part of the story. However, the decline relative to the DOW really began in August 2005 and continued right through December 2006. However, this gloomy perspective is only on a relative basis, and the turn around may have already occurred in January 2007. Both the local presence and local headquarters stocks were consistently up in 2006 over their close in 2005. They simply did not keep pace with the rest of the DOW. However, the January 2007 data just posted suggested a major rebound in the local stocks last month. While both categories still lag the DOW, particularly the local presence jumped way up and is now just 150 points below the full DOW. Next quarter we will address the relationship between the stock price index and local economic performance to address the causality and quality of local stocks and retail sales and unemployment.
The Borrom Line
Overall, 2006 was a good year to be in Jacksonville. Retail sales were way up, unemployment was very low, and inflation was virtually nonexistent. Local companies may not have seen as much upward movement in their stocks as would be optimal, but 2007 is starting out quite positively. If the housing market rebounds as many expect during at least the second quarter of 2007, and oil prices remain relatively stable, we see no evidence of a significant downturn in 2007 locally, and in fact strong economic growth for entire year.
Jacksonville’s Super Bowl: Economic Impact on Retail Sales
What does the Super Bowl represent? To the teams that play in the game the Super Bowl represent at least one full year of hard work and determination. For the fans, the Super Bowl is a celebration of sorts, family and friends get together, and good times are shared by all. For the lucky city that gets to host the spectacle that surrounds the game, the Super Bowl represents their chance to show the entire world the beautiful sites, the local scenery, and the positive characteristics of their fair city.
A Super Bowl committee is assembled each year to decide which lucky city will get to host the “ultimate football game” in the future. The committee usually designates which fortunate city will get the honor of holding the game four years in advance. In November of 2000 the Super Bowl committee voted to let Jacksonville hold the prestigious event, despite the fact that Jacksonville has the smallest total population of any community with an NFL football team. Most people in Jacksonville, however, were elated that the most important game in sports to most Americans would be played in their own backyard.
The NFL estimated that the Super Bowl would bring to Jacksonville an estimated 250 - 400 million dollars. Previous economic impact studies of games in Houston, Atlanta, and Miami suggested economic impacts of upwards of $300 million. The bulk of this revenue was to come from the influx of outside resources into the overall Jacksonville business community associated with hotel rooms, restaurant purchases, travel and entertainment locally, etc.
If these estimated revenues came to pass, then one would expect that there was a substantial increase in retail sales over the entire North Florida area between 2000 and February of 2005 when the game took place. Retail sales are a measure of local business activity, and are based on a survey of firms inside the desired sample area. Studies of previous Super Bowl events in recent years have failed to identify major increases in retail sales during the months either preceding or just following the game. So, was Jacksonville like these other cities, or did we see the direct impact of major increases in retail sales leading up to and during our Super Bowl?
The majority of Jacksonville’s local government officials, and economists, would have to agree ex post, that the Super Bowl brought no where near this amount of money into the local economy observable through retail sales prior to the game. Sometimes an event, such as the Super Bowl, can trigger a change in underlying spending habits of residents, or put into motion a set of events that can go unseen, at least on the surface of the overall economy. Could the Super Bowl have been such an event? We can test this hypothesis with statistical methods to try and ascertain the actual impact of the 2005 Super Bowl on the local Jacksonville economy.
Once we correct for cyclicality and seasonality in the retail sales data for Duval County a fairly stable trend starts to emerge. If an event like the Super Bowl were to instantaneously bring an influx of 250 - 400 million dollars to the Jacksonville economy, one would expect a very sharp upturn in retail sales, at least for the period preceding and during the Super Bowl fortnight. This sharp turn would result in what statisticians call a “break” in the series. This break could be easily observed and we could determine whether or not the break was statistically significant at some given level of statistical probability.
If the major increase in retail sales occurred, we would expect to see a break in the above series at possibly two specific points. The first point would be November of 2000 when the Super Bowl committee announced that Jacksonville would be the host city. The other point would be in February 2005 when the Super Bowl actually took place.
In November of 2000 when the Super Bowl committee announced its decision the overall change in retail sales for the month of November was actually down about five percent from the preceding year. Retail sales were actually down almost eight percent from the preceding year in December as well, so any appearance of a lag in the data is not likely. In January of 2005 retail sales were up about five and a half percent from the previous year. However, the overall change from year to year was only somewhere in the range of six million dollars. This type of increase is hardly what the NFL implied with their estimates.
These findings are bolstered by a survey conducted by the University of North Florida which stated that 76 percent of the businesses surveyed received no additional business from the Super Bowl. So it seems that the Super Bowl did not really aid the Jacksonville economy that much at all – at least from the standpoint of retail sales, preceding the event. However, deeper inspection of the data does reveal a statistically significant influence subsequent to the game. There is a definitive change in the trend of the retail sales function since January 2005. The graph below illustrates the point.
There seems to be a distinct change in the slope of this function somewhere around the thirty seventh time period. This particular graph was constructed from a retail sales data set starting in 2002, implying that the change that took place corresponds to January, 2005. Specifically, statistical tests confirm at the five percent significance level that there is a unique upward adjustment in the slope of the trend function subsequent to the thirty seventh time period of this data set (corresponding with January of 2005.)
Consequently, it appears that the major impact of the Jacksonville Super Bowl is reflected in more steady long term growth resulting from the events surrounding the Super Bowl rather than directly preceding or during the game. Because of the Super Bowl took place in Jacksonville, roads were restored or widened, and the entire city seemed to get a much needed “face lift.” This attracted new businesses and generated a boost in consumers’ personal income. This boost in business and consumer income is the factor reflected in the trend in retail sales over the last two years.
So, how big is this post-Super Bowl impact? By extending the trend in retail sales in the area before the Super Bowl in the two years since, and comparing that trend to the post-Super Bowl trend, we can isolate numerically both how much higher retails sales are in Jacksonville now compared to what they would have been had the previous trend continued; and by computing the area of the triangle determined by the two trends between January 2005 and February 2007, we can infer the total effect. Contrary to the discouraging results prior to the Super Bowl, the improvement in the trend since the game suggests that retail sales this month will be over $39 million higher than they would have been without the Super Bowl, and the triangle area implies that the total impact in the intervening two years has been almost $980 million. Naturally, there may be other factors that caused the increase in the retail sales trend that we cannot control for, but the post-Super Bowl impact may in fact be significantly more than what the NFL predicted.
It is very hard to determine how long this trend will continue before the multiplier effect of the Super Bowl wears off. But, it is reasonable to anticipate that if Miami receives much impact from the 2007 Super Bowl, it will not be directly reflected in short term increases in pre-game retail sales, rather in the structural improvements in the Miami community preceding Super Bowl Sunday that will continue in the form of increased business activity in subsequent years. The impact in Miami may not be as large as that which occurred and is occurring in Jacksonville because Miami has hosted Super Bowls before and thus did not require the infrastructure improvements that Jacksonville generated. However, regardless of the inert influences on local retail sales leading up to the 2005 Super Bowl, few could legitimately argue that the Super Bowl did not positively influence Jacksonville’s standing as a major city in the United States.
The Jacksonville MSA Phillips Curve
One of the most empirically sound relationships in macroeconomic theory is the postulated short run inverse relationship between the rate of inflation in a given economy and the level of unemployment. A simple example illustrates the implications. Everything else equal, as the nominal wages of workers increase firms will be motivated to increase the prices charged for goods and services to maintain a given level of profit. If this outcome is sufficiently pervasive, the price level (the aggregate level of prices) will rise accordingly. This in-turn causes wage setters to revise their expectations about their prospective profitability and to set their nominal wages higher to compensate for the rising price level. The result is a spiral of declining employment with greater inflation. This scenario is often dubbed the wage-price spiral. Such an outcome presumes that the economy is not in recession and creating jobs. Job creation pushes the unemployment rate below what is called the natural rate of unemployment or the NAIRU (non-accelerating inflationary rate of unemployment), or the level where inflation and the unemployment rate are both at their equilibrium values.
If an economy is not generating employment growth, the roles are reversed. Unemployment will be above the natural rate, and generally inflation will be low, ceteris paribus. Recovery will imply job creation, the inflation rate will rise, and the unemployment rate will spiral downward towards its equilibrium level.
The empirical evidence supporting this theory was first developed by British economist A. W. Phillips in the 1950s. Phillips plotted the unemployment rate for the United Kingdom against the level of wage inflation, and found a consistent inverse relationship between the two. Subsequent studies throughout the world and during different time periods continued to reflect similar outcomes. However, more recent studies have definitively rejected that there is a corresponding long run inverse relationship. Rather, the data indicate a vertical representation at the NAIRU, suggesting that in the long run the movements are up and down relative to inflation, with unemployment eventually returning to the NAIRU after short run adjustments.
During the LEIP projects nearly five year tenure we have been collecting data on several economic indicators in the Jacksonville area. Two of these economic time series are the CPI (consumer price index.), and the level of unemployment in the local economy. Thus, we can construct our own version of the much celebrated Phillips curve.
Both theory and past empirical results suggest that the short run Phillips curve would be hyperbolic or quadratic in nature i.e.: (1/x), or x2. Part of the justification is that the relationship tends to be asymptotic in nature because unemployment and inflation are both bounded below by zero (although deflation very infrequently does occur). The long run Phillips curve is presumed to be vertical.
As the graph below suggests, the Jacksonville economy follows this predicted pattern for the four plus years of data available (a period sufficiently long to reflect the long run).
Consequently, we can estimate the natural rate of unemployment for the Jacksonville MSA.
Of what value is this task? If one knows what the rate of unemployment is at a particular time, an estimate of the likely impact on inflation can be derived. This can assist local businesses and government entities in applications from investment to automobile purchases, and even housing refinance. If you know that the natural rate of unemployment is given by a and the unemployment rate is below a, you now know from reading this article that inflation will likely rise, and wages will have to drop so that the unemployment level rises and the economy reaches equilibrium. In other words you might want to hold off purchasing that second car for a little while or investing in that new plant.
From this research, we can deduce that within a ninety percent confidence interval, the current natural rate of unemployment for the Jacksonville area is somewhere between 1.58% to 3.2% (this is derived from Fiellers method of significance)2 . Jacksonville’s persistent falling level of unemployment, which reached a low of 2.85% during May, 2006, and the stable rate of core inflation (core inflation excludes food and energy) generate this seemingly low level for the natural rate. It is very important to note, however, that even though this is what the natural rate of unemployment is right now, there is no assurance that the natural rate of unemployment will take this same value in a year, or even six months from now. As circumstances change within the economy, the natural rate of unemployment will change accordingly. However, over the last four plus years, the stability in the long-run Phillips relationship implies that this level may continue for some time.
- Blanchard, Oliver. Macroeconomics, Forth edition. Pearsons Prentice Hall. 2006
- Journal of economic perspectives, Volume 1- winter 1997, pages 33-49. “The NAIRU, unemployment, and monetary policy.” Dougles Straiger, James H. Stock, and Mark W. Watson.