Complex information and accounting standards: Evidence from UK narrative reporting

Figures & data

Table 1. Data collection.

Event	Firm Years
Initial annual report textual output for 2003, 2004, 2006, and 2007	1,505
Less: firm years that have data missing for the above 4 years	(497)
Less: firm years with missing/extreme values for lowtfidf scores (close checking indicated poor word conversion of pdf report)	(56)
Final narrative firm-year observations =	952
Less: firm years with missing financial statement data	(13)
Final firm-year observations (used in hypothesis test)	939

Table 2. Variable definitions.

Variable	Full Name	Definitions
numrepwords	Report Length	Number of words in the annual report
numcompwds	Report Complexity	Number of complex words in the annual report
comp words	Percent Report Complexity	[(Number of Complex Words in the Annual Report / Total Number of Words in the Same Report) *100]. Complex words are words with three or more syllables This is the component of the fog index of interest in this study. The fog index is defined as: Fog=0.4(words per sentence + percent of complex words).
hightfidf comp words	High TF.IDF Words	Number of high TF.IDF complex words in the annual report
hightfidf	Percent High TF.IDF	Percentage of total words in the annual report that are high TF.IDF complex words
lowtfidf comp words	Low TF.IDF Words	Number of low TF.IDF complex words in the annual report
lowtfidf	Percent Low TF.IDF	Percentage of total words in the annual report that are low TF.IDF complex words
length	Length	Log of the number of words in the annual report
year_ifrs	Year IFRS	Year IFRS is an indicator variable equal to 1 for post-adoption years (2006 and 2007) and 0 for pre-adoption years (2003 and 2004)
Control Variables
firm_size	Firm Size	Natural logarithm of a firm’s total assets
firm_age	Firm Age	Date in 2012 (02/07/2012) less date of incorporation
earn_vol	Earnings Volatility	Standard deviation of annual earnings (operating earnings after depreciation) of the 5 years prior to the year of analysis
price_vol	Price Volatility	Average annual price movement to a high and low price from a mean price for each year
bus_comp	Business Complexity	Natural logarithm of the number of business segments
geo_comp	Geographical Complexity	Natural logarithm of the number of geographical segments
earns	Earnings	Firm’s operating income after depreciation scaled by the total assets of the firm
time1	Time 1	Measures the increase in word complexity for the pre- and post-IFRS adoption years; increases by 1 for each year
time2	Time 2	Measures the increase in word complexity for the post-IFRS adoption years; takes the value of 0 for the pre-adoption years and increases by 1 for each post-IFRS adoption year
profit_loss	profit loss	An indicator variable where 1 = profit firms and 0 = loss firms

Table 3. Descriptive statistics.

Panel A: Whole sample
Variable	Obs	Mean	Std. Dev.	Min	Max
Narrative Variables
numrepwords	952	36012.27	24887.97	3323.00	224890.00
numcompwds	952	9104.21	6497.61	808.00	56550.00
comp words	952	25.12	1.42	4.50	29.99
hightfidf comp words	952	7314.40	5217.72	654.00	44381.00
hightfidf	952	20.16	1.20	3.68	23.83
lowtfidf comp words	952	1795.12	1317.44	154.00	12235.00
lowtfidf	952	4.98	0.66	0.82	7.58
fog	952	21.94	1.35	16.50	30.59
Control Variables
firm_size	931	6.80	2.07	1.70	14.46
firm_age	856	40.80	28.83	2.55	111.98
earn_vol	913	146.68	581.57	0.00	7090.75
price_vol	856	28.76	10.78	11.82	75.41
bus_comp	806	0.88	0.62	0.00	2.30
geo_comp	793	1.00	0.59	0.00	2.30
earns	929	0.08	0.09	−0.46	0.74
profit_loss	929	0.92	0.27	0	1
Panel B: Pre- and post-IFRS adoption periods
year_ifrs=0	Obs	Mean	Std. Dev.	Min	Max
numrepwords	476	28452.95	19615.83	3323.00	162584.00
numcompwds	476	7100.15	5039.08	808.00	40534.00
comp words	476	24.82	1.30	16.89	29.53
hightfidf comp words	476	5748.38	4051.95	654.00	31708.00
hightfidf	476	20.08	1.13	13.09	23.56
lowtfidf comp words	476	1355.90	1015.73	154.00	8883.00
lowtfidf	476	4.76	0.61	2.94	6.99
year_ifrs=1	Obs	Mean	Std. Dev.	Min	Max
numrepwords	476	43571.58	27217.97	3634.00	224890.00
numcompwds	476	11108.26	7147.37	1090.00	56550.00
comp words	476	25.42	1.48	4.50	29.99
hightfidf comp words	476	8880.42	5759.60	866.00	44381.00
hightfidf	476	20.24	1.26	3.68	23.83
lowtfidf comp words	476	2234.34	1434.10	205.00	12235.00
lowtfidf	476	5.19	0.64	0.82	7.58

Table 4. Correlation matrix – selected variables.

		1	2	3	4	5	6	7	8
comp words	1	1.0000
hightfidf	2	0.8871*** (0.0000)
lowtfidf	3	0.5461*** (0.0000)	0.0980** (0.0025)	1.0000
firm_size	4	0.0974** (0.0029)	0.1236*** (0.0002)	−0.0118 (0.7184)	1.0000
price_vol	5	−0.0478 (0.1620)	−0.0199 (0.5605)	−0.0684* (0.0454)	−0.2913*** (0.0000)	1.0000
bus_comp	6	0.0552 (0.1174)	0.0625 (0.0764)	0.0067 (0.8494)	0.3518*** (0.0000)	−0.1760*** (0.0000)	1.0000
earns	7	−0.1156*** (0.0004)	−0.0630 (0.0548)	−0.1354*** (0.0000)	−0.0355 (0.2794)	−0.2248*** (0.0000)	0.0008 (0.9829)	1.0000
year_ifrs	8	0.2105*** (0.0000)	0.0705* (0.0296)	0.3258*** (0.0000)	0.0915** (0.0052)	−0.1393*** (0.0000)	0.0459 (0.1926)	0.1104*** (0.0000)	1.0000

Parentheses denote significance: * < 0.05, ** < 0.01, *** < 0.001.

Table 5. Determinants of complexity.

		(1)	(2)	(3)	(4)
	Predicted Sign	Length	comp words	hightfidf	lowtfidf
firm_size	+	0.1665***	0.0070	0.0356	−0.0278
		(19.42)	(0.23)	(1.36)	(−1.90)
earn_vol	+	0.0001*	0.0003**	0.0001	0.0002***
		(2.04)	(3.05)	(1.49)	(3.92)
price_vol	+	0.0024	0.0008	0.0025	−0.0015
		(1.55)	(0.15)	(0.52)	(−0.57)
bus_comp	+	0.1124***	−0.0632	−0.0734	0.0108
		(4.60)	(−0.71)	(−0.98)	(0.26)
geo_comp	+	0.1596***	0.5785***	0.5806***	−0.0029
		(6.44)	(6.41)	(7.68)	(−0.07)
earns	+/−	0.5156**	−1.6804**	−0.5319	−1.1438***
		(3.11)	(−2.79)	(−1.05)	(−4.04)
_cons		8.8472***	24.5686***	19.3519***	5.2208***
		(100.64)	(76.69)	(72.16)	(34.76)
N adj. R^2		709 0.580	709 0.092	709 0.102	709 0.037
F-statistic		169.90 ***	17.74***	16.19 ***	5.54 ***

t statistics in parentheses.

* p < 0.05, ** p < 0.01, *** p < 0.001.

Table 5 presents results from estimating the determinants of the key dependent variables. See Table 2 for variable definitions. The regression analysis is performed with robust standard errors to obtain unbiased estimates under heteroskedasticity. We first present results of a model without firm fixed year effects in Table 5. Doing so permits us to investigate the impact of variables that exhibit little within-firm variation over our sample period. However, given that our controls variables have limited variables that exhibit these characteristics, we perform the remaining analysis with firm fixed effects.

Table 6. Narrative complexity and mandatory IFRS adoption.

	(1)	(2)	(3)	(4)
	Length	comp words	hightfidf	lowtfidf
year_ifrs	0.424***	0.718***	0.247**	0.470***
	(15.64)	(7.18)	(3.17)	(10.06)
firm_size	0.044	−0.138	−0.078	−0.062
	(1.44)	(−1.23)	(−0.85)	(−1.26)
earn_vol	0.000	−0.000*	−0.000	−0.000
	(0.56)	(−2.20)	(−0.79)	(−1.36)
price_vol	0.00202	0.000	0.000	−0.000
	(0.69)	(0.02)	(0.08)	(−0.14)
bus_comp	0.0657	−0.083	−0.043	−0.038
	(1.27)	(−0.45)	(−0.31)	(−0.51)
geo_comp	0.084*	0.348*	0.269*	0.079
	(2.16)	(2.33)	(2.04)	(1.20)
earns	0.116	−1.047	−0.369	−0.672*
	(0.64)	(−1.49)	(−0.62)	(−2.23)
Firm Fixed Effects	Yes	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes	Yes
N adj. R^2	709 0.608	709 0.127	709 0.021	709 0.308
F-statistic	84.88***	20.50***	4.09***	21.70***

t statistics in parentheses.

* p < 0.05, ** p < 0.01, *** p < 0.001.

Table 6 presents the results from estimating the regression of complexity variables and IFRS adoption. year_ifrs is an indicator variable equal to 1 for the post-IFRS adoption years and 0 for pre-adoption years. All control variables are included in the regression model. See Table 2 for variable definitions. The regression analysis is performed with robust standard errors to obtain unbiased estimates under heteroskedasticity.

Table 7. Narrative complexity and IFRS adoption with economic and time effects.

	(1)	(2)	(3)	(4)
	Length	comp words	hightfidf	lowtfidf
year_ifrs	0.265***	0.804***	0.247	0.560***
	(4.90)	(3.47)	(1.26)	(6.22)
time1	0.107***	0.087	0.133**	−0.046
	(8.71)	(1.88)	(3.29)	(−2.04)
time2	−0.0583*	−0.137*	−0.155	0.016
	(−2.54)	(−1.16)	(−1.58)	(0.41)
firm_size	0.023	−0.135	−0.085	−0.051
	(0.85)	(−1.42)	(−1.07)	(−1.06)
earn_vol	0.000	−0.000*	−0.000	−0.000
	(0.49)	(−2.03)	(−0.68)	(−1.30)
price_vol	0.004	0.001	0.002	−0.001
	(1.21)	(0.09)	(0.22)	(−0.25)
bus_comp	0.070	−0.081	−0.039	−0.040
	(1.38)	(−0.43)	(−0.27)	(−0.54)
geo_comp	0.087*	0.352*	0.274*	0.078
	(2.29)	(2.36)	(2.09)	(1.18)
earns	0.016	−1.066	−0.436	−0.623*
	(0.09)	(−1.46)	(−0.70)	(−2.02)
Firm Fixed Effects	Yes	Yes	Yes	Yes
Year Fixed Effects	Yes	Yes	Yes	Yes
N adj. R^2	709 0.628	709 0.126	709 0.023	709 0.308
F-statistic	72.24 ***	20.26 ***	4.39 ***	22.70 ***

t statistics in parentheses.

* p < 0.05, ** p < 0.01, *** p < 0.001.

Table 7 presents results from estimating the regression of complexity variables, including time1 and time2 variables as independent variables. time1 indicates the increasing complexity of annual reports over time, and time2 indicates increasing complexity in the post-IFRS adoption years. All control variables are included in the regression model. See Table 2 for variable definitions. The regression analysis is performed with robust standard errors to obtain unbiased estimates under heteroskedasticity.

Table 8. Earnings, business complexity, and IFRS adoption.

	(1)	(2)	(3)	(4)
	Length	comp words	hightfidf	lowtfidf
year_ifrs	0.416***	0.413	−0.056	0.469**
	(10.42)	(2.40)	(−0.43)	(5.90)
bus_obf	0.013	0.100	0.108	−0.007
	(0.35)	(0.99)	(1.23)	(−0.06)
earns_obf	−0.045	2.257*	2.169**	0.082
	(−0.20)	(1.98)	(2.45)	(0.44)
firm_size	0.043	−0.123	−0.064	−0.061
	(1.42)	(−1.14)	(−0.72)	(−1.27)
earn_vol	0.000	−0.000*	−0.000	−0.000
	(0.54)	(−2.29)	(−1.17)	(−1.37)
price_vol	0.002	−0.005	−0.004	−0.001
	(0.74)	(−0.33)	(−0.36)	(−0.17)
bus_comp	0.059	−0.123	−0.088	−0.034
	(1.08)	(−0.71)	(−0.66)	(−0.45)
geo_comp	0.086*	0.284*	0.208	0.076
	(2.10)	(2.30)	(1.93)	(1.13)
earns	0.153	−2.502**	−1.761*	−0.730
	(0.57)	(−2.81)	(−2.20)	(−1.85)
Firm Fixed Effects	Yes	Yes	Yes	Yes
Time Fixed Effects	Yes	Yes	Yes	Yes
N adj. R^2	709 0.608	709 0.133	709 0.032	709 0.306
F-statistic	90.23 ***	18.84 ***	5.46***	17.76 ***

t statistics in parentheses

* p < 0.05, ** p < 0.01, *** p < 0.001

Table 8 presents results from estimating the regression of the complexity variables and the independent variables bus_obf and earns_obf, which are the interaction of year_ifrs with business complexity and earnings, respectively. All control variables are included in the regression model. See Table 2 for variable definitions. The regression analysis is performed with robust standard errors to obtain unbiased estimates under heteroskedasticity.