It's no wonder family-run upscale department store chain

Dillard's

(DDS) - Get Report

isn't making any new friends. It's having a tough enough time keeping its old ones.

After four consecutive months of positive same-store sales growth in 2004, the company reported a negative May comp last week. That came as somewhat of an unpleasant surprise, and it dashed already meager hopes for an operational turnaround that had appeared just a few weeks earlier, after the company said it had more than doubled its first-quarter profit.

That first-quarter bonanza, though, was in stark contrast to the previous fiscal year, when the company posted losses in two quarters and reported a profit in another quarter that missed the consensus by more than 30 cents a share.

While Dillard's earnings inconsistencies and reluctance to communicate with Wall Street aren't new trends, they continue to make it difficult to analyze the company. The company's annual shareholders meeting in May, for example, reportedly lasted 20 minutes with 10 people in attendance. Chief Executive William Dillard II spoke for just five minutes.

Thus comparing quarterly and annual results with analysts' expectations can be misleading, especially because the company does not provide its own quarterly earnings estimates. In fact, the company is so reticent that most analysts and portfolio managers don't feel comfortable talking about it.

What's the Problem?

But that doesn't mean there are a lack of opinions. "There's no fundamental story to tell. Their operating results are disgraceful," said a buy-side analyst who wished to remain anonymous. "Dillard's is really stuck in the early '90s. They haven't adapted, they haven't changed."

Company sales fell to $7.6 billion in fiscal year 2003, down from $7.9 billion in 2002 and $8.2 billion in 2001.

The waning sales stem not only from a slowness to adapt more fashionable brands but also from instituting less price discounting than its peers, analysts say.

These factors have effectively been a consistent drag on same-store sales. In 2003, same-store sales -- largely seen as a gauge of a retailer's overall health -- fell in 10 out of 12 months.

As a result, the company earned just $9.3 million in fiscal 2003, a huge drop-off from a year earlier, when the company posted a profit of $131.9 million before a charge due to an accounting change.

Despite posting a mini-turnaround in comps during the first four months of the year, the company said June 3 that May same-store sales dropped 5% vs. the consensus expectations for a 1.8% increase.

"Dillard's surprisingly weak

May sales suggest that market share continues to erode, particularly in the more moderate segment," said UBS analyst Linda Kristiansen in a recent research note.

At the core of its dysfunction, according to the unnamed buy-side analyst, is that the Little Rock, Ark.-based company's top management team is manned by at least three Dillard family members, who are essentially sucking the company dry because they either don't care about the business or don't have the retail flare to run it. (The Dillard family owns about 6% of the company and can re-elect two-thirds of the board, since they hold 99% of the company's Class B shares.)

Falling on Deaf Ears

For all the disparagement, though, the company's shares are up nicely, rising 45% year over year to around $20. They are currently outperforming competitors

Federated

, which is up 34%, and

May

(MAY)

, which has risen about 20% in the past 52 weeks.

Amidst a mostly strong first quarter for the rest of the retail industry, Dillard's shares bounced after its own positive first-quarter earnings report, sailing to a new 52-week high of $20.96 on June 2, the day before news of the May same-store sales decline.

But the blowout first quarter didn't fool Wall Street analysts. The results do not "reflect any sustainable shift in Dillard's fundamentals in view of merchandising and real estate challenges," wrote Kristiansen in a May 13 research note. At the time, the analyst expected more market share loss and said the strong earnings are likely not sustainable.

"Dillard's reported numbers tend to swing broadly vs. consensus," said Banc of America analyst Dana Cohen. "While the first quarter was strong, this company has far exceeded Street numbers on a few occasions over the past few years, but has been equally likely to disappoint." (Banc of America does investment banking for Dillard's.)

And it's really the real estate value of the company that moves the shares, according to the buy-side analyst. The company's property holdings have long been known to be a value play, because the company owns many of its locations -- mostly in malls -- rather than leasing them. Kristiansen values Dillard's real estate as worth $18 a share.

However, for all the charges to the contrary, the company actually does have a merchandise strategy. During the first quarter, Dillard's said it presented merchandise with a younger focus and more fashion-forward attitude. The company said it is also offering selections from promising new upscale national brands. (A telephone call to Dillard's was not returned.)

"Like many of the traditional department stores, Dillard's is aiming to shift its mix from more moderate towards 'better,' introducing some higher-priced private label brands," said Kristiansen. (UBS does investment banking for Dillard's.)

Dillard's also is starting to enhance its exclusive brand program to include more selections in the "better" categories. Dillard's will continue to de-emphasize or eliminate underperforming brands from both national brand and exclusive brand sources.

Its private label business, which makes up about 21% of merchandise, includes the brands Kate Landry, Daniel Cremieux and Antonio Melani.

Kristiansen noted, however, that Dillard's is still behind in its more high-fashion approach. Federated, for example, began private-label brand development eight years ago.

In the end, it might take the management of the company changing familial hands again to create any positive momentum for this company. That could occur only "if they say they want to maximize this thing

and need these shares to be worth something," said the buy-side analyst.