- John Polk said “I knew Charles when he was EVP of The Atlanta Chamber and I worked for ...” on Memories of Oklahoma City circa 1993
- John Polk said “Back in the mid-80's and early 90's, Cleveland was actually recognized as one of the ...” on Economic development in NEO: A view from the street-level
- John Polk said “Is there any way to substantiate Dimora's claim re: GCP and the PD, other than ...” on Cleveland’s new development dynamic?
- George Nemeth said “Like all glimmers of newness in CLE+ I expect this one to be crushed too” on Cleveland’s new development dynamic?
- Cleveland’s new development dynamic? | Brewed Fresh Daily said “[...] by Ohio voters, as gambling interests convert the Ohio constitution into a zoning ordinance. ...” on Ohio’s casino deal gets a bit more messy
- About BDP Comments
The Plain Dealer published some devastating RTA ridership numbers last weekend. If 2009 plays out as expected, RTA will have its lowest year of ridership since the agency was formed in the mid-70s.
Last month I argued here that RTA fares are among the highest in the country, which led some to ask me how RTA fares compare to themselves historically. Obviously we know that in dollars and cents, fares are higher than ever, but what really matters is what they look like in inflation-adjusted dollars.
The blue line on this graph shows the nominal fares since October 1975 – or the fare that you would pay out of pocket. The red line shows the inflation-adjusted fares (using base period 1982-84).
In other words, not only are RTA fares the highest they have ever been in real dollars, but in the static period between October 1975 and September 2009, fares more than doubled overall inflation.
I went ahead and built a model to see what relationship exists between RTA ridership and a few logical variables. What influences ridership? I think fare prices are obvious, regional population seems intuitive, and gasoline prices are often cited as a reason why people use transit instead of driving. Lastly, it’s often suggested in the media that transit usage dips during recessions because if people don’t have jobs, they don’t have anywhere to go – so I threw in unemployment for good measure.
The model points to both predictable and surprising patterns (you can download my datasets and the regression output here, if you’d like).
Real Fare Prices: As the real fares go up, ridership goes down. Ridership is negatively correlated with real fare price (-.51).
Population: As Cuyahoga County population shrinks, ridership declines along with it. The two variables are positively correlated (.74).
Gasoline Prices: Although high gasoline prices in the past few years have led to headlines about crowded buses and trains and stories about people giving up driving, my model shows only a nominal increase in ridership relative to historical values (refer back to the ridership graph above to understand why). RTA ridership, in fact, was highest during years when gasoline was historically inexpensive. The decades-long trend points to a negative correlation between gasoline prices and ridership (-.40) and the past few years as statistical outliers.
Unemployment: My model finds a positive correlation between ridership and unemployment (.63). But it should be noted that the relationship is not statistically significant to the model.
If I toss out unemployment, 94% of the variation in ridership can be explained by the model (R-sq = .94). There is one important variable that I did omit, but not because I wasn’t thinking about it – the level, frequency, and quality of transit service. The reason this isn’t in the model is because it is difficult to measure and good data that does exist does not go back far enough in time. My hunch is that service cuts have the same relationship to ridership as fare hikes – as service levels go down, ridership declines too.
Here’s the bad news: RTA really has control over only one of the variables in question. Population in Cuyahoga County is on a downward trajectory with no apparent end in sight. Gasoline prices may very well rise again as the economy recovers, which could have another small positive impact on ridership, but it will also force RTA to deal with the fact that the agency is a much bigger purchaser of motor fuel than the individuals switching away from cars because of fuel costs. RTA will surely face pressure to raise fares again in the future. It’s clear that without reforming the system’s funding mechanism, balancing annual budgets will remain an issue for the system at least in the near future. All else equal, fare hikes may be a quick budget fix, but we shouldn’t be surprised if they are accompanied by continued falling ridership.
Update (11/10/09): A follow-up to this post and discussion of the missing service-level independent variable can be found here.
Last 5 posts by Rob Pitingolo
- The Cleveland International Film Festival: Thinking Ahead - March 29th, 2010
- Omitted Variable Bias - February 18th, 2010
- Deciphering Another Nonsense Forbes List - February 10th, 2010
- A Downtown Cleveland Fantasy - December 23rd, 2009
- Put a Nail in RTA's Coffin - November 17th, 2009