Fair Value Measurements

6 Months Ended

May 31, 2013

Fair Value Disclosures [Abstract]

Fair Value Measurements

Fair Value Measurements

Recurring Fair Value Measurements

The following table details the fair value measurements within the fair value hierarchy of our financial assets and liabilities at May 31, 2013 (in thousands):

				Fair Value Measurements Using
	Total Fair Value			Level 1		Level 2			Level 3
Assets
Money market funds	$	47,911		$	47,911	$	—		$	—
State and municipal bond obligations	37,417			—		37,417			—
Auction rate securities – municipal bonds	23,562			—		—			23,562
Auction rate securities – student loans	2,938			—		—			2,938
Foreign exchange derivatives	(40		)	—		(40		)	—
Liabilities
Contingent consideration	$	379		$	—	$	—		$	379

The following table details the fair value measurements within the fair value hierarchy of our financial assets at November 30, 2012 (in thousands):

				Fair Value Measurements Using
	Total Fair Value			Level 1		Level 2			Level 3
Money market funds	$	125,591		$	125,591	$	—		$	—
State and municipal bond obligations	50,818			—		50,818			—
Auction rate securities – municipal bonds	23,420			—		—			23,420
Auction rate securities – student loans	2,901			—		—			2,901
Corporate bonds	2,607			—		2,607			—
Foreign exchange derivatives	(186		)	—		(186		)	—

When developing fair value estimates, we maximize the use of observable inputs and minimize the use of unobservable inputs. When available, we use quoted market prices to measure fair value. The valuation technique used to measure fair value for our Level 1 and Level 2 assets is a market approach, using prices and other relevant information generated by market transactions involving identical or comparable assets. If market prices are not available, the fair value measurement is based on models that use primarily market based parameters including yield curves, volatilities, credit ratings and currency rates. In certain cases where market rate assumptions are not available, we are required to make judgments about assumptions market participants would use to estimate the fair value of a financial instrument.

The valuation technique used to measure fair value for our Level 3 assets, which consists of our ARS, is primarily an income approach, where the expected weighted average future cash flows are discounted back to present value for each asset. The significant unobservable inputs used in the fair value measurement of our ARS are the probability of earning the maximum rate until maturity, the probability of principal return prior to maturity, the probability of default, the liquidity risk premium and the recovery rate in default. Generally, interrelationships are such that a change in the assumptions used for the probability of principal return prior to maturity is accompanied by a directionally opposite change in one or more the following assumptions: the probability of earning the maximum rate until maturity, the probability of default and the liquidity risk premium. The recovery rate in default is somewhat independent and based upon the ARS' specific underlying assets and published recovery rate studies.

One issuer of our ARS is currently in default, but the underlying bond insurer is making interest payments on the issuer's behalf. In this situation, we used a market approach, where the significant unobservable inputs are the market credit default swap spread and the credit rating of the insurer.

The following table provides additional quantitative information about the unobservable inputs used in our Level 3 valuations as of May 31, 2013:

	Valuation Technique	Unobservable Input	Range (Weighted Average)
Auction rate securities	Discounted cash flow	Probability of earning the maximum rate until maturity	0.3% - 9.4% (1.6%)
		Probability of principal return prior to maturity	79.1% - 95.2% (88.6%)
		Probability of default	4.0% - 12.8% (9.8%)
		Liquidity risk premium	4.0%
		Recovery rate in default	50% - 70% (56.5%)
	Market valuation	Market credit default swap spread of insurer	2.8%
		Credit rating of insurer	AA-

Significant increases or decreases in the underlying assumptions used to value the ARS could significantly increase or decrease the fair value estimates recorded in the consolidated balance sheets.

We have classified contingent consideration related to the Rollbase, Inc. acquisition (Rollbase), which occurred in May 2013 (Note 7), within Level 3 of the fair value hierarchy because the fair value is derived using significant unobservable inputs, which include discount rates and probability-weighted cash flows. We determined the fair value of our contingent consideration obligations based on a probability-weighted income approach derived from probability assessments of the attainment of certain milestones. We establish discount rates to be utilized in our valuation models based on the cost to borrow that would be required by a market participant for similar instruments. In determining the probability of attaining certain milestones, we utilize data regarding similar milestone events from our own experience. On a quarterly basis, we will reassess the probability factors associated with the milestones for our contingent consideration obligations. Significant judgment is employed in determining the appropriateness of these key assumptions as of the acquisition date and for each subsequent period.

The key assumptions as of May 31, 2013 related to the contingent consideration for the acquisition of Rollbase used in the model are probabilities in excess of 95% that the milestones associated with the contingent consideration will be achieved and a discount rate of 4.8%. A decrease in the probabilities of achievement could result in a significant decrease to the estimated fair value of the contingent consideration liability.

During the three months ended May 31, 2013, there was no change in the fair value of the contingent consideration liability related to the Rollbase acquisition.

The following table reflects the activity for our financial assets measured at fair value using Level 3 inputs for each period presented (in thousands):

	Three Months Ended					Six Months Ended
	May 31, 2013		May 31, 2012			May 31, 2013			May 31, 2012
Balance, beginning of period	$	26,442	$	33,343		$	26,321		$	33,539
Redemptions and repurchases	—		—			(25		)	(225		)
Transfer to Level 2 fair value measurement	—		(2,700		)	—			(2,700		)
Unrealized gains included in accumulated other comprehensive loss	58		805			204			834
Balance, end of period	$	26,500	$	31,448		$	26,500		$	31,448

During the second quarter of fiscal year 2012, we received a redemption notice for one of our ARS at par value. We transferred the ARS to a Level 2 fair value measurement, as the value at May 31, 2012 was based on observable inputs.

Nonrecurring Fair Value Measurements

The following table details our nonrecurring fair value measurements at November 30, 2012 (in thousands):

			Fair Value Measurements Using
	Total Fair Value		Level 1		Level 2		Level 3		Total Losses
Disposal group	$	16,487	$	—	$	16,487	$	—	$	8,601

The disposal group included the assets and liabilities held for sale of the Artix, Orbacus and Orbix product lines, which had a fair value of $16.5 million as of November 30, 2012. The carrying value at November 30, 2012 of $25.1 million was written down to fair value, less costs to sell, resulting in a loss of $8.6 million. The loss was recorded in income (loss) from discontinued operations in fiscal year 2012. The assets and liabilities held for sale were divested in the first quarter of fiscal year 2013 (Note 6).

We evaluate all of our assets held for sale using undiscounted cash flow models or other valuation models, such as comparative transactions and market multiples, to determine their fair values. However, a market approach was more heavily used to value the assets held for sale related to the divestitures of our non-core product lines. As bid and transaction values became apparent as we moved through the marketing and divestiture process, the fair values of the assets held for sale was established. The impairment loss recorded for the Artix, Orbacus and Orbix product lines was primarily based on our expectations of a sale price as compared to our estimation of the net assets to be sold at closing.