IS SUSTAINABLE SUPPLY GROWTH IN THE GOLF INDUSTRY'S FUTURE?
After last issue's departure from our analytic roots into the pure editorial mode, we'll return to a more balanced attack of facts and opinions in trying to make some sense out of the interesting pattern of golf supply growth. A headline in Golfweek.com last week declared 'Course openings dropped 13% in 2002' and cited industry information sources as saying The downturn had to come sooner or later' suggesting from both parties that this was an unwelcome thing and a negative indicator for the industry. We'd like to suggest that, in our view of the world, the headline should have read 'Course openings return to a more sane, sustainable level in 2002' and the industry commentary should have said We welcome this healthy correction from unsustainable levels of new supply and believe that it will set the stage for an improving overall golf economy in the upcoming years for operators and lenders alike.'
Those of you who have been with us since September of last year might remember that in the 9/18/02 issue we outlined the case that, in a balanced world, annual demand increase patterns didn't require any 300 new course years for the next five years. That said, our 2002 call to maintain 'annual equilibrium' was 276 courses. What that means is that in our view of the world, the industry took the first step towards correct supply-demand imbalances by not overbuilding against 2002's projected demand increase (don't cheer yet, there's still the '95-01 excess inventory to absorb). In this context is the support for our modified headline above of returning to sustainable levels vs. a 'disappointing' drop.
In the wake of such a considerable one-year correction, it's often wise to look back and determine what might have caused the preceding boom and how we might avoid it in the future. Our discussions over the past 6 months with industry leaders suggest three primary causes for the boom:
1. A continued stream of 'aspiratio nal' golf course owners and investors who are driven more by dreams and ego than solid business principles regarding supply, demand and facility 'mix' that a given geography can support
2. The continued trend of major real estate developers to offer a golf course 'amenity' to developments funded by lot premiums and an exit strategy to divest vs. operate the facilities as viable stand-alone businesses (not saying this is a bad strategy, the financials seem to work for the developer but not for the eventual acquirer of the course or the other competitive courses in its draw area)
3. Poor or misinformation on the consumer fundamentals needed to support a successful facility starting all
the way back in the feasibility assessment of both new and existing operations
As an information provider to the golf industry we have little control over either of the first two potential causes. That said, we'll focus our attention on how information providers should be helping interested industry stakeholders make better-informed decisions on the consumer aspect of supply and demand both on a macro and local market basis.
What's wrong with the existing consumer approach? Having had the opportunity to read and review about a dozen feasibility and market studies over the past year, several common themes emerge that we believe can and should be improved upon:
o Consumer facts supporting key assumptions are often 2 or more years old
o Participation and frequency are being 'proxied' from broad geographies such as regions
o Demographic weighting is improperly applied randomly crossing RH measures (like HH Income) with population facts (like participation)
o Demographic weighting (like participation by Age) is being applied to local markets using national weights
o Consumer and facility-reported rounds are being compared to determine 'potential'
o Varying levels of precision exist in determining the current 'demand' (which we would call facility-reported rounds) for the facility's 'draw area'
o No factual basis for determining a facility's 'draw area'
Taking a one-sentence approach, we'll briefly touch on the potential hazards of each point in isolation. First, due to the fact that golf participation information has historically not been consistent or updated annually (perhaps driven in part by the market's unwillingness to pay for current information?), it is not unusual to see baseline numbers from these broad geographies, when they're sourced, to date back 2-3 years in age. Second, while there is almost universal industry consensus that 'all golf is local', the majority of participation and frequency facts used to calculate golfer and rounds information are regional facts covering areas as broad as 9 states, 3 climate zones and participation variances among those states of +1- 50%. Third, existing calculations of golfers and rounds weighting by Population Age and HH Income often fail to factor in that current industry source participation data is the 12+ population universe (you need to strip out the under 12 pop as well as the fact that you lose any rounds the under 12 crowd are actually playing) and the fact that the participation numbers by HH income are not HH-based (how do we know, because national HH participation is roughly 17% and the values across the income groups as currently reported by industry sources average to the population participation number of 11.7% not 17%). Fourth, as we've previously outlined, a major flaw in current 'rounds potential' calculations is the apples-to-oranges comparison of consumer-reported rounds against facility-reported rounds (can you say, 'guaranteed upside', we knew you could). While we've acknowledged that our approach of collecting annual rounds played from facilities from our licensed course database has its own flaws, we're not willing yet to concede that in-market facility interviews or phone surveys produce the real answer (perhaps two different versions of liar's poker?). Fifth, it's curious to us how many facilities have a '10 mile' draw area for analysis with little substantiation as to the facts supporting that conclusion which is a key variable in determining how many consumers are needed to support the facility.
Other than these things, the current approach is fine but we think we've outlined a reasonable argument why poor facts and insight may have been a contributor to the recent supply boom. We don't have the perfect solution but we have the next generation of improved solutions. Over the past 3 years, we've struggled with solutions to the above challenges within the constraint of economically collecting golf information and have made advances on several fronts. First, we've completed sizeable national annual surveys for 2001 and 2002 which allow us to make state and sub-state information available on a consistent basis to address limitations of the first two points. With the introduction of GolfStats prior to the Orlando show, those interested in state-level participation and frequency have an economical source for better base facts for their independent analyses. Regarding the confusion between correctly doing population vs. HH weighting, we've attempted to make that process easier as well by giving parties two solutions: One, the ability to purchase HH Income, Age of Population and HH Income/Age of Head of Household weighting on a state-by-state basis and two, an 'outsourced' option of using our Golf Local Market Analyzer web-delivered tool to calculate it for any geography in the country.
Rather than try to explain it, let's look at an example below. In this analysis, we were asked to provide information regarding the consumer fundamentals for several potential 'draw areas' around a given facility. While we looked at 5, 10, 15 and 20 mile radii, we'll just walk through the 10-mile radii analysis here for brevity (see, I told you every course had a 10 mile draw area). Using consumer survey information at the 'sub-state' level, we put unweighted participation at 8.4% and rounds per golfer at 23.4 per year. The projected consumer-reported rounds from this baseline for the 10-mile radius is shown in the first column below. In the next three columns, we weighted that annual rounds projection by HH Income (yes, we used HH participations instead of Pop participations), Population Age (yes, we counted all the age groups and used less-than-national weightings) and finally by HH Income and Age of Head of Household (you can't do Income by Age because Income is a HH-based measure and Age is a Population-based measure). These four figures provide a reasonable range of annual rounds values that the consumer base should support for this area and allows the client a choice regarding which approach they believe produces the best answer or some averaged value (probably the appraiser's favored method). The way that we would interpret the three weighted values in relation to the baseline calculation would be: We've got an area with similar HH Income to the baseline but a younger population mix which skews our Pop Age projection downward and is strong enough to also pull down the HH lncomeIHOH Age calculation.'
The second chart below outlines what we believe is an improved approach to determining a consumer-reported rounds potential index. True potential is determined by taking some benchmark performance (here the value from the unweighted, sub-state area) and comparing the weighted values of the subject area looking for favorabilities or unfavorabilities introduced by the local population (and HH) variances. Comparing the annual rounds projection for each of the weighted scenarios back against the unweighted version, we've produced the indices in the chart below. The interpretation of this chart is: 'Introducing the 'draw area's' unique demographics produces similar and slightly lower projections than the unweighted analysis leading us to believe that this particular area has no upside in rounds production from the consumer fundamentals.' In summary, in this example we've introduced three improvements to the current process: We used sub-state vs. regional consumer facts, we properly matched HH facts to HHs and Population facts to Pops and we've used an apples-to-apples comparison of weighted to unweighted rounds projections from the same source (consumer survey) to produce a more accurate potential measure.
One final point of clarity here, this analysis doesn't necessarily say that you can't build another facility here, it simply implies that if one does, that the consumer fundamentals suggest that the rounds will more likely come from cannibalization than 'latent demand'. I had an objection to this type of analysis two years ago by one of the senior Textron Financial folks who asserted that We've built where this type analysis said 'don't' and succeeded while conversely building where it said 'build' and failed miserably. It doesn't work'. It's taken us almost two years to figure it out but what we believe he was missing is that he had no way of determining whether his facility had added to or cannibalized the consumer base so he was simply relying on the skills of the operator to win or lose in the market. When he picked a winning operator, they in some cases were able to steal consumers and win in an environment that had no upside consumer demand. On the flip side, when he chose a marginal operator, even a good consumer base couldn't make that project a success. What we learned is that the consumer analysis is but one small piece of making a successful investment and a single factor that tells one in advance whether the consumer fundamentals will work for or against the success of that particular project. Hopefully what they've learned as an organization is that picking loans based solely on the operator's credentials is likewise an inherently riskier proposition than a well-rounded analysis.
Should there be a separation between 'information' and 'insight' providers? This brings us to the final anomaly in the whole current process of improved information and insight upon which sustainable supply might be built going forward. We've had innumerable industry people in the feasibility and market study segment point out the strange situation in which they compete directly against the same organization which provides their underlying information. It seems to us that doing superior work in analyzing a golf property or market requires knowledge in three disparate areas: The consumer fundamentals, the current supply dynamics and the operational/financial benchmarks for successful facilities. We understand the limitations of our core competencies in consumer work and analysis and frankly believe that a better end-client (course owners, lenders and investors) solution is a complementary vs. competitive marrying of skills. Whether or not we're right on this front only time will tell but the marketplace does have a way of sorting out this type of dissonance overtime. The most common frustration to date from those recognizing this situation is We have no choice.' You now have a choice.
If you've made it this far, you're a dedicated OTR reader. In summary, we've outlined our case as to why the precipitous drop in course openings should be viewed as a positive indicator for the industry. We've also outlined our first take on how the industry might be able to make positive adjustments going forward to minimize the contribution to 'irrational exuberance' in the next supply buildup from the misinformation component outlined above. We've taken our approach from 'theory' to real-world application and outlined how it improves upon the current process on several dimensions. As always, we're open to comments, criticisms and kudos. For those of you with friends and acquaintances in the operator/ownerllender/appraiserlinvestor community, feel free to pass this along as a gentle nudge to 'think different' (like Apple Computer, Inc.).
For those of you attending the upcoming NGCOA conference in Palm Springs there are still openings on Jim's dance card for lively discussion and debate. Contact Jim directly (847.808.7651 or email@example.com if you're interested. Next up will be the Crittenden conference in Phoenix with the same offer for any of you attending that conference as well. Finally, Jim will be speaking at the Urban Land Institute golf conference in Naples, FL the 25th of March on a panel discussing 'Growing the Game: What Is Being Done and Is It Working?' We've just received the 2002 consumer survey data and the toplines show that the participation rate changed only marginally last year (some surprise, but won't say yet if it was up or down). Those interested in attending and hearing the details of our take on the consumer franchise changes in 2002 should contact the ULI (www.uli.org for information. We continue to advance ideas and hypotheses to improve consumer insight for golf's 'early adopters' most of whom we value as readers setting the industry standards by looking at things from 'Outside The Ropes'.
(,Copy gh1 2003 Pe ucid Cop. A ights 'ese'ved. Cuola1's rl penned. Mateia may not be 'eg'o&ced ,n v#oe 0' pal in any loin v,tasoeve. vAlhaJl witten consent of Pe ucid Cop.