Hedging, hedge accounting, and earnings predictability

Abstract

Studies suggest that, pursuant to the implementation of SFAS 133, even sophisticated users of financial statements find it difficult to comprehend earnings implications of hedging derivatives. Moreover, due to stringent hedge accounting requirements under these standards, many economic hedges do not qualify for hedge accounting and are deemed “ineffective” for financial reporting purposes. Motivated by these considerations, we investigate the impact of hedging on earnings predictability by analyzing hand-collected hedging data from two industries that extensively use derivatives to manage price risks: the oil-and-gas exploration and production industry and the airline industry. In contrast to extant evidence, we find that overall hedging derivatives improve income predictability and increase (decrease) analysts’ forecast accuracy (dispersion). We also show hedges deemed ineffective for hedge accounting can increase earnings volatility and significantly impair earnings predictability. This finding lends support to concerns expressed by some corporate managers and industry experts against stringent hedge accounting requirements.

Appendices

Appendix A

1.1 Hedge accounting and hedge ineffectiveness under SFAS 133

Prior to the adoption of SFAS 133, hedging derivatives were kept off the balance sheet. Once the manager determines that a given derivative is used for hedging (as opposed to trading), its impact is recorded on the income statement in the same period in which the underlying hedged transaction is recognized. In the event the derivative is not a perfect hedge, the impact of the imperfection is also recognized in the same period that the hedged transaction is recognized. Hence, pre-SFAS 133, to the extent that the derivative provided some economic protection against the price volatility of the hedged item, the income statement was shielded from this volatility.

SFAS 133 significantly altered derivative accounting by requiring that all derivatives be recorded on the balance sheet at fair value. Under SFAS 133, only hedges that qualify as “effective hedges” for accounting purposes are allowed to be accounted under hedge accounting, in which changes in fair value are recorded in other comprehensive income (OCI) until the hedged item is recorded in earnings (at which time these changes are also released to earnings). Therefore, the income statement effect of cash flow hedges that qualify for hedge accounting under SFAS 133 is broadly similar to that of pre-SFAS 133 period.

SFAS 133 imposes stringent guidelines for derivatives to be deemed effective hedges and qualify for hedge accounting. Assessments of hedge effectiveness are required at the inception of the transaction, at least every three months thereafter, and whenever financial statements or earnings are reported (SFAS 133, ¶20b). While the standard does not prescribe a specific method of determining hedge effectiveness, most firms use the “80–125 dollar offset ratio standard,” in which the dollar offset ratio is defined as the change in the value of the hedging instrument divided by the change in the value of the hedged item over the assessment period. In other words, while a hedging instrument the value of which changes 75 cents in the opposite direction for every one dollar change in the price of the hedged item would substantially (albeit imperfectly) offset the underlying price risk, it would not qualify as an effective hedge for hedge accounting purposes and would therefore be deemed ineffective. Moreover, even highly effective hedges might be disqualified from hedge accounting in the absence of formal documentation of the hedging relationship and the entity’s risk management objectives and strategy (SFAS 133, ¶20a).

For hedges that do not qualify for hedge accounting (ineffective hedges), periodic fair value changes are reported directly in the income statement, regardless of when the hedged item is recognized in the income statement. This income statement recognition of interim changes in derivative fair values exposes the firm to potential earnings volatility (accounting effect), even though the economic objective of the hedge is to reduce cash flow volatility (economic effect).

At the end of this appendix, we provide a stylized illustration of how failure to qualify as effective hedges under SFAS 133 could provide economic and accounting outcomes that conflict with one another.

Many firms find it difficult to meet the high threshold of hedge effectiveness stipulated by SFAS 133 (i.e., the 80–125 dollar offset ratio or the 80%–125% negative correlation), due to factors such as differences between the basis of the hedging instrument and the hedged item as well as differences in certain terms of the hedging instrument and hedged item, such as maturities, quantity, location, or delivery dates. For example, an oil-and-gas firm might find it difficult to obtain derivatives that meet the hedge accounting standards, because the price reference location of highly liquid exchange-traded crude oil and natural gas derivatives may not fully correspond with their production location.³¹ Similarly, since jet fuel derivatives are not exchange traded, many airlines use crude oil and heating oil contracts as economic substitutes. While these substitutes would significantly counter jet fuel price risks, they may fail to meet SFAS 133 effectiveness thresholds and hence may not qualify for hedge accounting. Moreover, as a practical matter, many firms hold a combination of both effective and ineffective hedges. Two illustrative disclosures relating to airlines using heating oil contracts as substitutes for jet fuel hedges and firms holding both effective and ineffective hedges are presented in Appendix B.

As explained previously, critics contend that high effectiveness thresholds and onerous documentation requirements imposed by SFAS 133 to qualify for hedge accounting cause the potential unintended consequences of hedging derivatives creating artificial income volatility making earnings less predictable.

1.2 Illustration of Economic and Accounting Outcomes of Hedge Effectiveness

This illustration considers hedging outcomes for a hypothetical crude oil producer XYZ, wanting to hedge the sale of 10,000 barrels of crude oil taking place at the end of four reporting periods from now with forward contracts. We consider three scenarios.

• a. Entire sale is hedged with a fully effective forward contract that qualifies for hedge accounting (Tables 6 and 7)

• b. Entire sale is hedged with a contract that is an economic hedge but one that does not qualify for hedge accounting (Tables 8 and 9)

• c. Half the sale is hedged with the contract in (a) above while the remaining half is hedged with the contract in (b) above (Table 10)

The following assumptions are made.

• Fixed price stipulated in forward contracts is $50 per barrel. Hence the expected value of the transaction is $500,000 (10,000 x $50).

• XYZ neither receives nor pays an upfront premium so that the fair value of the derivative at inception is zero.

• Time value of money is ignored in all fair value calculations.

• Spot price per barrel for the type of crude oil produced by XYZ turns out to be $44.00, $40.50, $39.30, and $38.10 at the end of periods one, two, three, and four respectively.

1.3 Scenario a: A fully effective hedge that qualifies for hedge accounting

Table 6 This table shows the crude oil spot price per barrel, fair value change during the period, and cumulative fair value change for both the derivative and the anticipated hedged transaction at the end of each period. Note that the periodic price changes in the derivative and underlying hedged transaction are perfect (i.e., −100%), implying a fully effective hedge

Table 7 This table shows the income statement, statement of other comprehensive income (OCI), and cash flow effects of the derivative and the hedged transaction at the end of each period

As can be seen in column (10) of Table 7, the net cash flow effect of derivative and the hedged transaction is a cash inflow of $500,000 at the end of period four, which secures an effective price of $50.00 per barrel. The income statement effect (column 4) is identical to the cash flow effect, because interim fair value changes of the derivative are recognized in other comprehensive income (OCI) and released back to the income statement when the hedged transaction occurs at the end of period four. With this forward contract, XYZ effectively eliminates price uncertainty of period four crude oil production from both cash flow and income statement standpoints. Thus, effective hedging dampens variations in both cash flows and earnings by offsetting fluctuations in output and factor prices.

1.4 Scenario b: An economic hedge that does not qualify for hedge accounting

Table 8 This table shows the crude oil spot price per barrel, fair value change during the period, and cumulative fair value change for both the derivative and the anticipated hedged transaction at the end of each period. Note that, unlike in Scenario a, even though fair value changes of the derivative instrument and the underlying hedged transaction are negatively correlated, they are not perfect. From a practical standpoint, this can be viewed as a case where the actual production location and the production location on which the forward contract is based on differ. The correlation between the fair value changes in the derivative and the underlying hedged transaction is −75%. Under extant hedge accounting rules this hedge would be deemed ineffective and would not qualify for hedge accounting.

Table 9 This table shows the income statement, statement of other comprehensive income (OCI), and cash flow effects of the derivative and the hedged transaction at the end of each period

Referring to column (10) of Table 9, the net cash flow effect of the derivative and hedged transaction is a cash inflow of $506,000 at the end of period four, which secures an effective price of $50.60 per barrel. In other words, the derivative has significantly (albeit imperfectly) shielded XYZ from price fluctuations in the crude oil market. However, because the derivative does not qualify for hedge accounting, interim changes in the fair value of the derivative are recognized not in OCI, but directly in the income statement. Thus, the effective price reported in period four when the hedged transaction takes place is $39.10 per barrel. Consequently, the income statement is exposed to fluctuations in crude oil prices and fails to reflect the fact that the price risk was substantially hedged. This illustrates that, while economic hedges can significantly reduce risks associated with price volatility, to the extent that they do not qualify for hedge accounting, such risk reductions are not reflected in the income statement.

1.5 Scenario c: A combination of the fully effective hedge that qualifies for hedge accounting and the economic hedge that does not qualify for hedge accounting

Table 10 This table shows the income statement, statement of other comprehensive income (OCI), and cash flow effects of the derivative and the hedged transaction at the end of each period under a scenario where half the sale is hedged with the contract in (a) above while the remaining half is hedged with the contract in (b) above. This can be thought of as the common situation of a firm holding both effective and ineffective hedges

Under this scenario, interim fair value changes of the derivative that qualifies for hedge accounting do not impact the income statement and are recognized in OCI instead. On the other hand, interim fair value changes of the derivative that do not qualify for hedge accounting are directly recognized in the income statement. Column (10) of Table 10 reveals that the net cash flow effect of the derivative and hedged transaction is an inflow of $503,000 at the end of period four, which secures an effective price of $50.30 per barrel. The income statement impact lies in between scenario a. and b. above, resulting in an income of $445,500 in period four, reflecting an effective price of $44.55 per barrel. Holding both types of derivatives (ones that do and do not qualify for hedge accounting) provides some protection to the income statement against price risk in crude oil prices, but it is smaller when compared with the economic protection provided in terms of cash flows.

Appendix B

1.1 Examples of Hedge Ineffectiveness Disclosures

The following disclosure made by US Airways Group, Inc., in its 2007 10-K filling illustrates the issue of using heating oil contracts to hedge against jet fuel price risk and the consequent origination of hedge ineffectiveness.

The Company utilizes financial derivative instruments primarily to manage its risk associated with changing jet fuel prices. The Company currently utilizes heating oil-based derivative instruments to hedge a portion of its exposure to jet fuel price increases. These instruments consist of costless collars. As of December 31, 2007, the Company has entered into costless collars to hedge approximately 22% of its 2008 projected mainline and Express jet fuel requirements. The Company does not purchase or hold any derivative financial instruments for trading purposes. (p. 79)

In 2007, US Airways realized operating income of $524 million and income before income taxes of $485 million. Included in these results is $245 million of net gains associated with fuel hedging transactions. This includes $187 million of unrealized gains resulting from the application of mark-to-market accounting for changes in the fair value of fuel hedging instruments as well as $58 million of net realized gains on settled hedge transactions. US Airways is required to use mark-to-market accounting as our existing fuel hedging instruments do not meet the requirements for hedge accounting established by SFAS No. 133, “Accounting for Derivative Instruments and Hedging Activities.” If these instruments had qualified for hedge accounting treatment, any unrealized gains or losses would have been deferred in other comprehensive income, a component of stockholder’s equity, until the jet fuel is purchased and the underlying fuel hedging instrument is settled. (p. 44)

-US Airways Group, Inc. 10-K, 2007

The following disclosure by Forest Oil Corp. in its 3Q 2001 10-Q filling provides a typical example of a firm employing a combination of both effective and ineffective hedges.

(O)n January 1, 2001, the Company began accounting for the energy swaps and collars, in accordance with SFAS No. 133. All of Forest’s energy swap and collar agreements and a portion of Forest’s basis swaps in place at January 1, 2001 have been designated as cash flow hedges. As a result, changes in the fair value of the cash flow hedges are recognized in other comprehensive income until the hedged item is recognized in earnings, and any change in fair value resulting from ineffectiveness is recognized immediately in earnings. Changes in the fair value of basis swaps not designated as cash flow hedges are recognized in other income.³² The increase in fair value of derivative financial instruments included in other comprehensive income during the third quarter and nine months ended September 30, 2001 was $14,308,000 and $30,774,000, respectively. … Included in other income during the third quarter and nine months ended September 30, 2001 are net gains (losses) of $(8,806,000) and $2,298,000, respectively, on basis swaps and other instruments not designated as cash flow hedges.

-Forest Oil Corp 10-Q, 3Q-2001

Appendix C

1.1 A model of hedge accounting

In this appendix, we model the income statement effects of hedging by a firm in steady state to analytically demonstrate how hedge accounting affects earnings volatility.

Let $\tilde{x}_{u}$ represent period-specific fluctuations (i.e., gain or loss) of a going concern’s exposure to price risk of the underlying item in steady state. By definition, $E\left(\tilde{x}_{u}\right)=0$. Let $Var\left(\tilde{x}_{u}\right)={\sigma}^2$. Let a fraction θ (0 < θ < 1) of hedging positions closed every period because of the culmination of the underlying transaction. For a firm to be in steady state, we assume that new exposures of the same magnitude originate in each period so that the full extent of exposure remains unchanged from period to period.

The firm wishes to hedge this exposure with a derivative instrument. Let period-specific fluctuations in the market value of hedging derivative portfolio be represented by $\tilde{x}_{d}$ with $E\left(\tilde{x}_{d}\right)=0$. Let $Var\left(\tilde{x}_{d}\right)={\sigma}^2$. We assume the same variance for the two random variables for analytical convenience but without loss of generality.³³

Let $\mathrm{Cov}\left({\overset{\sim }{\mathrm{x}}}_{\mathrm{u}},{\overset{\sim }{\mathrm{x}}}_{\mathrm{d}}\right)={\uprho}_{\mathrm{u}\mathrm{d}}{\upsigma}^2$, where ρ_ud is the standard Pearson correlation coefficient. For the derivative portfolio to be a hedge it must be that case that ρ_ud < 0 . The net exposure in any period is then given by

$$ {\displaystyle \begin{array}{c}{\tilde{\mathrm{x}}}_{\mathrm{net}}={\tilde{\mathrm{x}}}_{\mathrm{u}}+{\tilde{\mathrm{x}}}_{\mathrm{d}}, with\\ {}E\left({\tilde{x}}_{net}\right)=0, and\\ {}\begin{array}{c} Var\ \left({\tilde{x}}_{net}\right)= Var\left({\tilde{x}}_u\right)+ Var\left({\tilde{x}}_d\right)+2 Cov\left({\tilde{x}}_u,{\tilde{x}}_d\right)\\ {}=2{\sigma}^2\left(1+{\uprho}_{\mathrm{u}\mathrm{d}}\right).\end{array}\end{array}} $$

With this structure, observe that $Var\ \left({\tilde{x}}_{net}\right)=0$ for ρ_ud =  − 1, reflecting a perfect hedge. In general, for hedging to be economically beneficial, it must be the case that

$$ {\displaystyle \begin{array}{c} Var\ \left({\tilde{x}}_{net}\right)< Var\left({\tilde{x}}_u\right)\\ {}\iff 2{\sigma}^2\left(1+{\uprho}_{\mathrm{ud}}\right)<{\sigma}^2\\ {}\iff {\uprho}_{\mathrm{ud}}<-\frac{1}{2}\end{array}} $$

We will henceforth assume that all hedges satisfy this condition—otherwise there is no purpose in hedging.

We define what constitutes an ineffective hedge from an accounting perspective in reference to a threshold $\overline{\rho}\upvarepsilon \Big(-1,-\frac{1}{2}$), to capture the essence of SFAS 133.

Definition 1 Given a threshold $\overline{\rho}\upvarepsilon \Big(-1,-\frac{1}{2}$), a hedging derivative is deemed ineffective from an accounting perspective if ${\rho}_{ud}>\overline{\rho}$; it is deemed effective otherwise; i.e., if ${\rho}_{ud}\le \overline{\rho}$.

Thus hedging is economically beneficial but ineffective if ${\uprho}_{\mathrm{ud}}\upvarepsilon \left(\overline{\uprho},-\frac{1}{2}\right)$. On the other hand, hedging is economically beneficial and effective if ${\uprho}_{\mathrm{ud}}\upvarepsilon \left(-1,\overline{\uprho}\right]$.

Note that, in practice, firms typically apply the negative 80% correlation threshold to determine whether a derivative qualifies as an effective hedge for accounting purposes. Therefore, in general, firms likely have some effective and some ineffective hedges. To motivate our hypotheses of how effective and ineffective hedges might affect income volatility, we next consider two extreme cases—when the entire hedging derivative portfolio is fully effective and when it is fully ineffective.

• Case 1: Fully effective hedging derivative portfolio: $-1<{\uprho}_{\mathrm{ud}}\le \overline{\uprho}$

In this case, hedging will affect income statement volatility only when the underlying hedged item is “consummated” which happens for a fraction θ of the exposure every period.³⁴ Consequently, hedging will decrease income statement volatility if

$$ Var\ \left(\theta {\tilde{x}}_{net}\right)< Var\left(\theta {\tilde{x}}_u\right) $$

$$ \iff \kern1em Var\ \left({\tilde{x}}_{net}\right)< Var\left({\tilde{x}}_u\right) $$

This inequality is always satisfied when ${\uprho}_{\mathrm{ud}}<-\frac{1}{2}$. Moreover, because ${\uprho}_{\mathrm{ud}}\le \overline{\uprho}<-\frac{1}{2}$ for effective hedges, effective hedges always decrease income statement volatility.

• Case 2: Fully ineffective hedging derivative portfolio: $\overline{\uprho}<{\uprho}_{\mathrm{ud}}<-\frac{1}{2}$

In this case, the net income is also affected by immediate recognition of gains and losses associated with the ineffective hedging derivatives associated with the (1 - θ) fraction of hedged item that remain unconsummated at the end of the period. Accordingly, the net income effect in a period may be computed as

$$ {\tilde{\mathrm{x}}}_{\mathrm{net}}=\uptheta \left({\tilde{\mathrm{x}}}_{\mathrm{u}}+{\tilde{\mathrm{x}}}_{\mathrm{d}}\right)+\left(1-\theta \right){\tilde{\mathrm{x}}}_{\mathrm{d}}. $$

It follows that

$$ {\displaystyle \begin{array}{c} Var\ \left({\tilde{x}}_{net}\right)={\theta}^2 Var\left({\tilde{\mathrm{x}}}_{\mathrm{u}}+{\tilde{x}}_d\right)+{\left(1-\theta \right)}^2 Var\left({\tilde{x}}_d\right)+2\theta \left(1-\theta \right) Cov\left({\tilde{\mathrm{x}}}_{\mathrm{u}}+{\tilde{x}}_d,{\tilde{x}}_d\right)\\ {}=2{\theta}^2{\sigma}^2\left(1+{\uprho}_{\mathrm{u}\mathrm{d}}\right)+{\left(1-\theta \right)}^2{\sigma}^2+2\theta \left(1-\theta \right)\left[ Cov\left({\tilde{\mathrm{x}}}_{\mathrm{u}},{\tilde{x}}_d\right)+ Var\left({\tilde{x}}_d\right)\right]\\ {}\begin{array}{c}=2{\theta}^2{\sigma}^2\left(1+{\uprho}_{\mathrm{u}\mathrm{d}}\right)+{\left(1-\theta \right)}^2{\sigma}^2+2\theta \left(1-\theta \right)\left[{\uprho}_{\mathrm{u}\mathrm{d}}{\sigma}^2+{\sigma}^2\right]\\ {}=2{\theta \sigma}^2\left(1+{\uprho}_{\mathrm{u}\mathrm{d}}\right)+{\left(1-\theta \right)}^2{\sigma}^2\end{array}\end{array}} $$

The income statement volatility will decrease relative to when there is no hedging if and only if

$$ {\displaystyle \begin{array}{c}2{\theta \sigma}^2\left(1+{\uprho}_{\mathrm{ud}}\right)+{\left(1-\theta \right)}^2{\sigma}^2<{\theta}^2{\sigma}^2\\ {}\iff 2\theta {\uprho}_{\mathrm{ud}}+1<0\\ {}\iff \theta {\uprho}_{\mathrm{ud}}<-\frac{1}{2}\end{array}} $$

Because 0 < θ < 1 and ρ_ud < 0, the larger the θ, the more negative the ρ_ud, or both, the more likely the above inequality holds. This condition implies that the economically beneficial effects of hedging will overwhelm the accounting effects of ineffective hedging when a larger fraction of derivative positions is closed every period, price fluctuations of hedging derivatives are more negatively correlated with the underlying hedged item, or both. Conversely, accounting effect will overwhelm the economic effect and increase earnings volatility when a large fraction of derivative positions remains open, when the negative correlation between price fluctuations of hedging derivatives and the underlying hedged item is relatively weak, or both.

We recognize that the parameter θ, which represents the fraction of exposure for which the underlying hedged item is “consummated” every period, cannot be empirically estimated from available data, especially given that it likely varies across firms and varies over time for the same firm. Our motivation in presenting this model is to simply establish that ineffective hedges need not always increase income volatility because of how they are accounted for under SFAS 133. The economic effect associated with these hedges could well outweigh the deleterious accounting effect and result in a decrease in income volatility.

Appendix D

1.1 Computation of Extent of Derivative Usage (Derivatives)

1.1.1 Oil-and-Gas Industry

This example illustrates the computation of derivatives measure (Derivatives) for Stone Energy Corp. for second quarter of 2004.

As reported in its first quarter 2004 10-Q fillings, Stone Energy had the following outstanding derivative contracts to cover its second quarter 2004 production (that is, derivative contracts that become exercisable during second quarter of 2004).

1.1.2 Crude oil

Put options – 682,500 Barrels (Bbls)

1.1.3 Natural Gas

Swaps – 1,365,000 thousands of cubic feet (MCF)

Put options – 8,190,000 Mcf

According to the second quarter 2004 10-Q fillings, Stone Energy’s crude oil and natural gas production for the period were 1,552,000 Bbls and 14,443,000 Mcf respectively.

Therefore, for the second quarter of 2004, Stone Energy used derivative contracts to hedge 43.98% of its crude oil production (682,500/1,552,000) and 66.16% of its natural gas production [(1,365,000 + 8,190,000)/14,443,000].

The oil-and-gas industry uses a standard conversion rate of one Bbl of oil to six Mcf of gas to convert oil production into natural gas equivalent. Therefore 39.20% of Stone Energy’s second quarter 2004 total production relates to crude oil [(1,552,000*6)/{(1,552,000*6) + 14,443,000}], and the remaining 60.80% relates to natural gas [(14,443,000)/{(1,552,000*6) + 14,443,000}].

Applying these relative production fractions to derivative hedging fractions of crude oil and natural gas, it can be seen that, for the second quarter of 2004, Stone Energy used derivative instruments to hedge 57.46% of its total production [(0.4398 × 0.3920) + (0.6616 × 0.6080) = 0.5746].

Hence, the variable Derivatives takes the value for 0.5746 for Stone Energy in the second quarter of 2004.

1.2 Airline Industry

The following excerpt from United Airlines’ third quarter 2004 form 10-Q typifies the extent of hedging disclosures provided by most airlines.

“During the second quarter of 2004, we began to implement a strategy to hedge a portion of our price risk related to projected jet fuel requirements primarily through collar options. … Currently, we have hedged approximately 36% of our fourth quarter 2004 projected fuel requirements at an average price of $1.00 to $1.17 per gallon, excluding taxes.”

The company clearly states that 36% of expected jet fuel consumption is covered by derivative contracts. Hence, for the fourth quarter of 2004, the variable Derivative for United Airlines would take the value of 0.36.

Appendix E

1.1 Variable Definitions

Variable name	Definition
Accuracy	Analyst forecast accuracy: The absolute value of the difference between the individual analyst earnings forecast and the actual earnings scaled by stock price at the end of the quarter t for firm i. The values are multiplied by −100, so that greater values indicate more accurate forecasts. Specifically, |CEFit - EPSit|/Pit*(−100), where CEFit, EPSit, and Pit are the most recent individual analyst quarter earnings forecasts, actual earnings per share, and quarter-end price per share for firm i at period t, respectively.
Dispersion	Analyst forecast dispersion: The inter-analyst standard deviation of quarter earnings forecasts deflated by stock price at the end of quarter t for firm i. The value is multiplied by 100. Specifically, [SDit/Pit]*100, where SDit and Pit are the standard deviation of quarter earnings forecasts and quarter-end price per share for firm i at period t, respectively. Note that Dispersion changes as new analyst forecasts are issued.
Derivatives	The extent of derivative usage: For oil-and-gas industry, Derivatives is measured as the fraction of firm i, period t production covered by derivatives contracts. For airline industry, Derivatives is measured as the fraction of firm i, period t estimated jet fuel consumption covered by derivatives contracts.
Analysts	The total number of analysts following firm i at quarter t.
Size	Natural log of total assets of firm i at the beginning of quarter t.
Intangible	Intangible ratio: Ratio of intangible assets to total assets at beginning of year t.
Volatility	The standard deviation of monthly stock returns for firm i at year t-1.
MB	Market-to-book ratio: Market value of equity (prccq×cshoq) divided by book value of equity (atq-ltq-pstkl/4 + txditcq+dcvt/4).
Issue	An indicator variable taking the value of 1 if firm i issues equity greater than 5% of total assets at quarter t and 0 otherwise.
Turnover	Stock turnover: The ratio of the number of shares traded in quarter t to the average number of shares outstanding in quarter t.
Return	Market adjusted stock return: Quarterly stock return for firm i at quarter t-1, adjusted for contemporaneous quarterly market return.
ROA	Return on assets: Income before extraordinary items (ibq) divided by total assets (atq) at the beginning of quarter t.
Foreign	Foreign operations: An indicator variable taking the value of 1 if foreign income or loss (pifo) is not equal to 0 and 0 otherwise.
M&A	Mergers and acquisitions: An indicator variable taking the value of 1 if cash flow from mergers and acquisitions (aqcq) is not equal to 0 and 0 otherwise.
DA	Absolute value of discretionary accruals. The discretionary accruals is obtained using Jones (1991) accruals expectation model as modified by Dechow et al. (1995).
Ineffective	An indicator variable taking the value of 1 if firm i in quarter t have nonzero ineffective gain/loss and 0 otherwise.
AllIneffective	An indicator variable taking the value of 1 if none of the firms’ hedges qualify for hedge accounting and 0 if all of the firms’ hedges qualify for hedge accounting.
Num_firm	Number of firms analyst j follows in quarter t.
Num_ind	Number of industries analyst j follows in quarter t.
Brokerage_size	The size of brokerage that analyst j works in at quarter t, measured as the number of analysts working in the brokerage at quarter t.
Forecast_exp	Number of quarters analyst j has followed firm i.
Forecast_freq	Number of forecasts analyst j makes for firm i in quarter t.
Horizon	Number of days between forecast issuance date and actual earnings announcement date.