1. INTRODUCTION

Tomato cultivation and processing are present in various areas of Italy, where tomato is one of the most important food productions. Tomato processing firms require large amounts of capital to finance investment in fixed assets (FA), such as buildings and plants for tomato processing, and even to finance the cycle of net working capital (NWC). NWC absorption is particularly relevant in the sector due the increase in inventory stock and account receivable terms of payment. High absorption of FA and NWC, frequently financed with an increase in financial debt, makes it necessary to assess the sustainability of the business cycle in tomato processing firms. This assessment is particularly relevant in the actual situation of bank credit reduction and an increased number of firms' crises and insolvency. This analysis aims to provide useful information to managers in evaluating the firms' financial cycle sustainability. In recent years, even tomato processing firms were characterized by an increase in default rate; this is particularly true in the case of small and medium enterprises (SMEs) that generally have the worst access to capital markets and debt financing, as shown in several researches (Grablowsky, 1984; Dunn & Cheatham, 1993; Peel & Wilson, 1996; Molina & Preeve, 2009). To achieve these goals, this article analyzes the management data of a sample of tomato processing firms in Italy. The analysis considers firstly the balance sheet and income statement data, particularly with respect to capital structure, sustainability of the management cycle and credit access. The analysis considers firstly the balance sheet and income statement data, particularly with respect to capital structure, sustainability of the management cycle and credit access. The analysis also applies comparison between profit margins and financial margins to quantify statistically significant differences. Again, the article tests the differences for interest coverage ratios (ICRs) if calculated applying economic and a financial approach. Multiple regression models are tested at the end to quantify determinants of profit and cash flow in the tomato processing firms, as considered in the sample. The results of the research could be applied by policy makers through public actions in supporting private firms, even improving evaluation of financial viability of public aided firms and reducing risk of inefficient use of public aids (Boschi, Girardi & Ventura, 2014).

2. METHODS

Return on equity capital is defined as the increased value of equity capital in a given period and could be not only positive but preferably higher than the opportunity cost of capital (Lagerkvist & Andersson, 1996). Applying this point of view, return equity capital could be quantified with both economic and financial approaches. In an economic approach, revenues and costs are compared with the accrual-based methods, which quantify value creation, as expressed in accounting data. Financial approach -instead- considers cash inflow and outflow, applying a cash based approach that is suitable to quantify cash flow available to distribute dividends or to perform discretionary investments. Differences in firms' results after applying an economic or a financial approach are frequently caused by a lag between economic and financial cycles (Grenberg, Johnson & Ramesh, 1986; Kwon, 1989; Dechow, 1994; Dechow & Dichev, 2002; Russel, 2009; Iotti & Bonazzi, 2014). In fact, a traditional accounting system is based on the principles of historical cost and accrual basis value analysis; given the time lag between the economic and financial cycle, the importance of analyzing operating cash flows is expressed by several studies, particularly in regards to quantifying a firm's capacity to generate future cash flow (Finger, 1994; Wang & Eichenseher, 1998; Charitou & Panagitodes, 1999; Hussain & Al Attar, 2003). Again, several researchers have shown that economic and financial approaches have different results (Bowen & Owen, 1986; Kwon, 1989; Dechow, 1994), highlighting that firms may even have financial un-sustainability even in the case of positive income margins, both in the FA cycle (Fazzari & Petersen, 1993; Cleary, 1999; De Miguel & Pindado, 2001) and the NWC cycle (Howorth & Westhead, 2003; Padachi, 2006; Taylor, 2011; Baños-Caballero, García-Teruel & Martínez-Solano, 2014; Darun, Roudaki & Radford, 2015). Financial un-sustainability is particularly frequent in capital intensive firms as is often the case of agri-food processing firms (Glancey, 1998; Kieschnick, La Plante & Mousawwi, 2008; Sgroi, Di Trapani, Testa & Tudisca, 2014; Testa, Di Trapani, Sgroi & Tudisca, 2014).

2.1. ANNUAL ACCOUNT ANALYSIS

In the article, to quantify income and financial flows, annual account data is applied -according to the IV EU directive about company and annual accounts (EU, 1978)-, in Italy with the provision of the Italian civil code. The annual account contains a balance sheet, an income statement and integrative notes. For a given period, t, the balance sheet of a generic firm could be expressed as follows (Bonazzi, Iotti & Paduano, 2012):

In equation ⁽¹⁾, the left hand side represents investment, in which FA is fixed assets, WCia is working capital asset inventories, WCara is working capital asset accounts receivable, WCoa is working capital asset-other assets, and L is liquidity. The sources of capital are represented on the right hand side of equation ⁽¹⁾ where Esc is share capital, Er are reserves, PpT is profit after tax, WCapp is working capital debt accounts payable, WCop is working capital debt, other values, DFM<12 is financial debt due within 12 months and DFM>12 is financial debt due after 12 months. The left hand side of equation ⁽¹⁾ is total assets (TA), while the right hand side represents the total sources of capital, quantified as the total equity capital (ET = Esc + Er + PpT) and the total of debt capital (DT = WCapp + WCop + DFM<12 + DFM>12). Net financial position (NFPt) could be expressed as follows:

Net Working Capital (NWCt) expresses the absorption of capital as a result of the operating cycle, defined as acquisition, processing, and sale cycle:

In equation ⁽³⁾, in a given period t, WCaT is working capital total asset and WCpT is working capital total debt; NWC quantifies the net resources generated (NWCt < 0) or absorbed (NWCt > 0) by the working capital management cycle (Love, Preeve, & Sarria-Allende, 2007). NWC > 0 expresses a so called "conservative strategy" of working capital management because is verified (Hill, Kelly & Highfield, 2010) that NWC > 0 is inversely related to insolvency. NWCt < 0 expresses the case of "aggressive strategy" of working capital management (Grablowsky, 1984; Weinraub & Visscher, 1998) and is considered directly related to financial distress. Particularly in the tomato processing industry it is useful to express the reclassification of the balance sheet with the functional form considering NWC and NFP; in fact, tomato processing firms are often characterized by NWC absorption, to cover the cycle of raw material processing and the delay in accounts receivable payment. We could express it as follows:

In equation ⁽⁴⁾, FA and NWC, if positive, are covered by equity capital (ET) and net financial position (NFP); ET + NFP is then equal to net invested capital (NIC). The income statement then quantifies the accounting profit generated for equity holders. In the income statement, the value of production (VP) for a given period, t, is:

In equation ⁽⁵⁾, pt,i is the price per unit, at a given time t, of goods and services sold I : I ³ 1, qt is the quantity sold, and It,f and It"1,f are inventory items F : F ³ 1 and G : G ³ 1, respectively, at a given time, t and t-1. Every item is valued at its respective value per unit (v). St is total sales at a given time t, then DIt,t"1 is the variation in inventory values between t-1 and t. Inventories in the income statement and balance sheet have equal values, so at It = WCi^a_t and I_t-1 = WCi^a_t-1. The value of production (VP) is a flow value (Dechow & Dichev, 2002) that is in formation during a period, in our case t , without consideration of values during period T Î [t-1, t]. Operative costs, for a given time t, are as follows:

Mt is the cost of raw materials, St is the cost of services, Rt is the cost of rent and leasing, Lt is labor cost and Ot represents other operative costs. In equation ⁽⁶⁾, mt,m, st,s, rt,r, lt,l, and ot,o are the single factors of costs where M ³ 1, S ³ 1, R ³ 1, L ³ 1, O ³ 1; respective quantities are qt,m, qt,s, qt,r, qt,l, and qt,o. The nonfinancial operative costs are as follows:

In equation ⁽⁷⁾, Dt is amortizations of fixed assets (FAt), while At is depreciation; dt,d and at,a are the costs per unit, at a given time t, of amortizations and depreciation; these costs are, respectively, part of D : D ³ 1 and A : A ³ 1., with their respective quantities qt,d and qt,a. Balance of the financial operation (SFt), at t, is as follows:

In equation ⁽⁸⁾, I^a_t is interest receivable and I^p_t is interest payable at a given time, t. The balance of the extraordinary operations (SXt), at t, is as follows:

In equation ⁽⁹⁾, X^a_t is extraordinary component of income, and X^p_t is extraordinary expense. The balance of the management revaluations and devaluations of financial assets is given, at t, as follows:

In ⁽¹⁰⁾, Z^a_t represents revaluations while Z^p_t represents devaluations of financial activities. Income account could be then summarized as follows:

In equation ⁽¹¹⁾, EBITDA approximates the creation of liquidity, with an income approach before non-monetary costs (Dt + At); EBITDA even considers non-monetary values (DIt,t"1) of VP and thus does not directly express an operating cash flow margin but approximates it. EBIT is an income margin that expresses operative income after non-monetary costs (Dt + At); Π_t^aT is profit before taxes and Π_t^pTt (PROFIT) expresses the firm's capacity to remunerate equity capital; T^Y_t represents income taxes.

2.2. FINANCIAL STATEMENT ANALYSIS

Income statements are not suitable to use to conduct a financial analysis and then it is necessary to apply financial statements in quantifying available cash flow. The first definition quantifies cash flow as the sum of an accounting result (profit or EBIT) plus depreciation and amortizations (Beaver, 1966); other researchers began to express cash flow while taking into account the absorption or generation of cash via the working capital cycle (Rayburn, 1986; Wilson, 1987; Gombola, Haskin, Ketz, & Williams, 1987; Finger, 1994; Lorek & Willinger, 1996; Bradbury, 2011; Bond, Bugeja & Czernkowski, 2012; Clacher, De Ricquebourg & Hodgson, 2013; Farshadfar & Monem, 2013):

In equation ⁽¹²⁾, in the given period t, CFt is cash flow, OCFt is operating cash flow, UFCFt is unlevered free cash flow, FCFEt is free cash flow to equity, and NWC is net working capital. If ∆+NWCt,t"1 ≥ ∆"OCFt,t"1, an increase in net working capital implies an increase in absorption of liquidity, thus reducing OCF, and vice versa ("NWCt,t"1≥ ∆+OCFt,t"1). As it has been shown by several studies, NWC management is very important, especially for SMEs that frequently have limited access to the capital market in the medium and long terms and often finance-fixed assets with short-term liabilities (due within 12 months) and have rates of default higher than companies of larger sizes. The most part of tomato processing firms are, in fact, SMEs. Given OCFt, liquidity absorption due to FA investment has an effect on UFCFt, given that [(FAt " FAt"1) - (Dt + At)] > 0≥ ∆"UFCFt,t"1 and vice versa. UFCFt is therefore the cash flow available to serve NFP and ET. In a given period, t, FCFEt is the cash flow available for the distribution of dividends to equity holders. The aim of the research is to assess whether there are statistically significant differences in the income and financial results of firms in the sample, expressing income results in terms of EBITDA, EBIT, and PpT (profit) and financial results CF, OCF, UFCF and FCFE. In fact, when assessing the financial sustainability of the business cycle applying intermediate profit margins, such as EBITDA and EBIT (margins that approximate cash flow values), is necessary to consider that income margins do not consider the effect of revenue not collected from customers, purchases not paid to suppliers, and changes in the values of inventories, as NWC variation. Again, profit calculated with an income approach differs from cash flow available FCFE in consideration of the time lag between income value creation and the time of occurrence of financial flows. In the article we would compare income and financial margins in tomato processing firms' sample to verify the significance of any differences.

2.3. RATIO ANALYSIS

Income and financial margins are even applied to quantify interest coverage ratios (ICRs) that are calculated applying both a traditional approach (income approach), and a nontraditional (Bonazzi & Iotti, 2014, Iotti & Bonazzi, 2015) financial approach. Even in literature on bankruptcy, starting with Altman (1968), researchers try to explain financial distress by applying financial ratios, and the importance of ICRs in reducing asset volatility, when covenants are costly to enforce. This was expressed by several researchers (Leland, 1994, 1998). ICRs with an income approach are expressed as follows:

ICR1 and ICR2 express a firm's capacity to pay interest within a given period, t, with income margins (Healy, 1985; Sloan, 1996; Goldstein, Ju & Leland, 2001; Dothan, 2006; Moir & Sudarsanam, 2007). ICR2 expresses a more conservative approach: (Dt + At) ³ 0 ≥ EBITDAt ³ EBITt ≥ ICR2 ≥ ³ ICR1 as frequently used in bank loan agreements (Gray, Mirkovic & Ragunathan, 2006), when applying an income approach. Other authors (Bonazzi & Iotti, 2014) express ICRs for a given period, t, as follows ⁽¹⁴⁾:

ICR3, ICR4, and ICR5 express the capacity of the company to pay the cost of debt in a given period, t; in this case financial flow directly expresses the liquidity generated by the firm's cycle (CF, OCF, and UFCF). In the article, we would compare ICRs with income and financial approaches to assess if there are statistically significant differences. Again, in the article -to evaluate management's efficiency- profit and financial margins are applied to quantify return on equity capital. A first ratio is traditional return on equity (Rosner, 2003; Lewellen, 2004; Love, Preeve & Sarria-Allende, 2007), that is the most known accountancy measure of profitability:

In ⁽¹⁵⁾, ROE (return on equity) expresses the income annual return of equity capital (Cheng, Liu & Schafer, 1996; Ohlson, 1980; Barnes, 1987); ROE ratio is affected by accounting conventions, particularly related to the accrual principle; in fact, it is therefore possible to observe situations where firms, even in presence of positive profit, are not able to distribute dividends to equity holders. To overcome this problem, the work also proposes the application of the following ratio:

In ⁽¹⁶⁾, FOE (flow on equity) expresses the annual financial performance of equity capital (Bodenhorn, 1964; Moro Visconti, 1999). In the article, we compare ROEt and FOEt to quantify correlation and differences between ratios' values and even to suggest explication to ROE and FOE determinants.

3. DISCUSSION OF RESULTS

In 2013, about 40 million tons of tomatoes were processed worldwide. Italy, with 4.0 million tons of tomatoes processed, compared to about 68,000 hectares under cultivation, accounts for over 12% of the total world production and 55% of European production. The cultivation and processing of tomatoes characterizes various areas of Italy, a country in which the tomato is one of the most important components of agricultural food production. Italy is the second largest world producer of the tomato industry, second only to California. In recent years, the national tomato sector has been subject to profound changes, even considering that several emerging producer countries -including China, actually the third largest tomato producer- have increased their production with important changes in international trade dynamics, in both quantitative and qualitative terms (World Processing Tomato Council, 02/2014).

3.1. THE TOMATO SECTOR IN ITALY

In the period from 2009 to 2013, land surface for tomato production in Italy decreased by 11.68% and the contraction was greater for food consumption tomatoes, 15.12%, than for processing industry tomatoes, 10.01% (ISTAT, 2014); the production of tomatoes for food processing increased in the same period (0.98% in production and 1.24% in harvesting), with an increase in average yields per hectare, while the production of tomatoes for food consumption decreased by 16.50% in production and 16.72% in harvesting (ISTAT, 2014). Processing of tomatoes generally takes place in plants near the areas of production, particularly for cost reduction. Transportation costs reduction, in transporting raw materials from production areas to processing plants, have a great part in concentration process of tomato productions plants in Italy, which is particularly high in two main geographical areas.

The most important production area for tomatoes is located in the northern part of Italy, including the regions of Emilia-Romagna, Lombardia, Veneto, and Piemonte, which together produced 2.3 million tons in 2013 (44% of domestic tomato production). The area with the second highest production is located in the southern regions of Campania, Puglia, Calabria, and Basilicata, which together produced 2.2 million tons in 2013 (42% of national production). In southern Italy, 92 tomato processing firms were operating in 2013, of which 75 are limited companies, 11 are nonlimited companies and sole proprietorships, and 6 are cooperative firms. In northern regions, the tradition of processing tomatoes has its origins in the late 19th century with the rise of the canning industry in the province of Parma. This type of industrial process still characterizes the territories of the provinces of Parma and Piacenza, in the Emilia-Romagna region, with the presence of processing firms also in the Modena Province and in some other provinces of eastern Emilia-Romagna on the right side of the river Po and even in the lower areas of Lombardia (Cremona and Lodi Provinces), in the Veneto region, and in Alessandria Province (Piemonte region). In these territories, 22 processing firms are active, of which 16 are limited companies, 5 are cooperatives, and 1 is a non-limited company. Among the three major production regions of northern Italy, Emilia- Romagna prevails, with 15 processing firms. Some firms process only tomatoes, with production concentration in summer, employing seasonal workers during the production season; these firms frequently carry on maintenance, storage, and marketing activities during the rest of the year. Other firms cover a wide range of production, including canned vegetables and juices, with the aim to reduce seasonality of tomato production. Processing firms frequently suffer by low brand loyalty among consumers, thus reducing the bargaining power of processing firms. Processed tomatoes are largely sold to the consumer market by large retail chains that often use their bargaining power to apply unfavorable conditions to increase the average time of commercial credit, with an NWC cycle that determines an expansion of investment, even considering that sales to food distribution chains give a time increase in collecting receivables with increasing NWC financial absorption. Since NWC increasing has a positive impact on production value and profit, it is necessary to verify misalignment between profit margins and cash flow margins in the tomato processing firms industry to highlight situations wherein firms are not able to sustain financial cycle even with positive profit. Tomato processing firms require high investments even in FA (for buildings, plants, and equipment) with the need of an increasing financial source, in terms of ET or NFP investment, and this involves the need to quantify ICRs. Again, in recent years, tomato processing firms were affected by a large number of corporate crises, which have also led to bankruptcy and liquidation. In Italy, there has been a decrease in the number of firms in recent years, from 166firms in 2009 to 144 firms in 2013, with the closure of 22 firms, of which 11 are bankruptcies or compulsory liquidations.

3.2. FIRMS' SAMPLE DATA

The data considered in the analysis was made available by the register of companies, covering a five-year period, from 2009 to 2013. A sample of 50 firms with a total of 250 year-data is considered in the analysis. Data analysis was performed using the statistical package SPSS, issue 19.

The analysis of the sample firms first considers asset data 2009/2013 (Table Nº 1) of the annual balance sheet, which expresses capitalintensive activities for the firms' samples, given the value of production (TA/VP 132.43%). Capital intensive attitude is particularly due to the investment (44.52% of TA); tomato processing firms, in fact, often do investments in plants to process and store tomatoes during production season. Particularly important are investment equipment and highly technological machinery, as are lines of sterilization and packaging. Accounts receivable (WCara) are also one of the major voice of investment (25.55% of TA); again, high capital absorption is determined by large retail chains, whose bargaining power increases over an average number of days for credit payments (accounts receivable). To cover TA investment, firms in the sample use financial debts as the first source of capital, given that DFM<12 + DFM>12 is 42.95% of the TA, with a prevalence of loans maturing within 12 months (23.57% of TA), while loans maturing after 12 months are 19.37% of TA. Equity capital (ET) is the third source of capital (21.51% of TA), and it is lower than WCsT, which is equal to 35.53%. The analyzed capital structure shows that firms in the sector acquire a high debt level to finance investment, and a high level of financial debt increases financial dependence by the credit system, even increasing financial cost.

Table 1
Balance sheet data of tomato processing sample firms (2009/2013) Reclassification of the balance sheet with financial approach.

The analysis of the balance sheet in functional form (Table Nº 2) confirms the prevalence of investments in fixed assets (70.04% of NIC) and also highlights NWC > 231 cases out of 250, where the NWC mean is 29.96% of NIC. Asset characteristics of the sample firms express even high level of capital absorption in NWC cycle; in fact, the mean length of time to collect commercial credit as an account receivable (AR_DAYS) is 123 days (median value is 147); the mean length of time to pay commercial debt as an account payable (AP_DAYS) is 177 days (median value is 190), and the mean length of time for inventory stock (INV_DAYS) is 123 days (median value is 149). The average length of the operating cycle (AR_DAYS + INV_DAYS - AP_DAYS) is 75 days (median value is 175). Functional balance sheet analysis then confirms tomato processing firms' NFP dependence and consequently debt services sustainability evaluation. The income statements (Table Nº 3) quantify that the average value of VP amounts to € 12,131,548; in tomato processing firms the cost of raw materials absorbs the largest part of VP (45.80%). Cost of services is 19.97% of VP, while labor cost is 15.58% of VP as the thirdlargest cost; EBITDA and EBIT margin are 12.62% and 9.40% of VP respectively. The income statement shows that a relevant part of intermediate profit margins EBITDA and EBIT is absorbed financial charge, where balance of financial operation (SF) absorbs 4.56% of VP (i.e., 36.15% of EBITDA and 48.54% of EBIT). Net profit (PpT) is € 340,825 on average (2.81% of VP).

Table 2
Balance sheet data of tomato processing sample firms (2009/2013) reclassification of the balance sheet with functional approach.

Table 3
Income statement data of tomato processing sample firms (2009/2013) Reclassification of the income statement with value added approach.

The financial statement for firms' sample expresses in particular the cash flow absorption in the FA investment cycle. CF is 377.09% of PROFIT, and the average cash absorption of the NWC cycle amounts to € 170.701 (i.e., 50.08% of PROFIT), following that average OCF € 1.114.520 (327.01% of PROFIT). FA investment determines a positive UFCF for 150.60 % of PROFIT. In 195 out of 250 cases is UFCF > 0 and is UFCF > SF in 112 out of 250 cases, where it is not possible to have NFP reimbursement. The FCFE mean value is € 40.170 (median value is € 10.190) and is then FCFE > 0 in 112 out of 250 cases.

The analysis of financial statements (Table Nº 4) highlights some typical management characteristics of firms in the tomato food processing sector. The research particularly highlights that: 1) the dynamics of FA absorb a significant amount of liquidity generated by operations (OCF) as expressed by UFCF values; 2) UFCFs are not able, on average, to cope with the balance of financial operation (SF); 3) FCFE mean and median values are negative, thus highlighting the impossibility of dividend distribution and give NFP repayment.

Table 4
Financial statement of the tomato processing sample firms (2009/2013).

3.3. COMPARISON OF INCOME AND FINANCIAL MARGINS

Intermediate income margins EBITDA and EBIT results are higher than financial margins OCF and UFCF, respectively, considering both mean and median values. It is EBITDA > 0 in 241 cases, EBIT > 0 in 223 cases, CF > 0 in 249 cases, while OCF > 0 in 195 cases and UFCF > 0 in 179 cases. PpT (PROFIT) also has a value higher than FCFE; PROFIT is positive in 155 cases out of 250, while FCFE is positive only in 90 cases out of 250. The application of intermediate margins EBITDA and EBIT could overestimate the cash flow available, particularly if OCF and UFCF are considered the relevant financial flows (UFCF and OCF). The analysis thus highlights a remarkable difference between income and financial margins. In particular, FA investment absorbs a substantial amount of the liquidity of the sample firms as it is expressed by the median values of CF (€ 1.199.710), OCF (€ 1.084.564 ), and UFCF (€ 535.395). Meanwhile EBITDA (€ 1.410.985) and EBIT (€ 1.034.879) have median values higher then financial margins, and this shows a shift in the relationship between income and financial cycle; particularly in the FA investment cycle, as expressed by the median values of PROFIT and FCEE, which are € 306.996 and € -10.190, respectively. Income margin analysis could then give a distorted assessment of financial flows availability, given that even in the case of positive accounting remuneration of equity capital, sample firms are unable to generate cash flows available for equity holders. In order to analyze the relation between analyzed values, we first have to test the normality of the distribution by applying the Kolmogorov-Smirnov D (KSD) statistic for both, income and financial flow. The KSD evidences the non-normality of distributions for all considered ratios as expressed by several researches (McLeav, 1982; Barnes,1982; Ezzamel, Brodie & Mar-Molinero, 1987). Even considering the largely reduced number of observations, we apply a nonparametric approach to correlation (Spearman's r). The analysis (Table Nº 5) shows significant correlations between income margins and financial margins, with high significance (1.00%), particularly for EBITDA and EBIT margins as income values and CF and OCF as financial values. FA capital absorption for investment cycle makes UFCF statistically different from all other margins, both calculated with income and financial approaches. Again, the correlations between FCFE and other margins are not statistically significant.

Table 5
Correlation income and financial margins - nonparametric approach (Spearman's r).

Table 6
Comparison of income and financial margins nonparametric approach for paired samples (T-Wilcoxon).

To verify differences for mean values, a nonparametric approach such as the Wilcoxon statistic (Wilcoxon Matched-Pairs Signed-Ranks Test) for paired samples is applied, given the non-normality of the margin distributions. The analysis performs 7 comparisons, 3 of which are between EBITDA and financial margins (CF, OCF and UFCF), 3 between EBITDA and financial margins again (CF, OCF and UFCF), and 1 between PROFIT and FCFE. Pair wise comparisons show that in comparisons between EBITDA / OCF, EBITDA / UFCF, EBIT / UFCF, and PROFIT / FCFE we can reject the null hypothesis of equality between means with a two-sided test with significance at 1.00%. In comparisons between EBIT / CF we can reject the null hypothesis of equality between means with a two-sided test with significance at 5.00%. In comparisons between EBITDA / CF and EBIT / CF, we cannot reject the null hypothesis of equality between means with a two-sided test.

The analysis shows that income and financial margins are different even if they often correlated in the sample. EBITDA and EBIT are not statistically different from CF margins. In particular, tomato processing firms show absorption in the FA cycle; even the NWC cycle improves capital requirements, and this is particularly due to high average time in collecting commercial credits, particularly from large retail chains. In the tomato processing industry, management that considers only income margins could assume wrong decisions, particularly about firms' cycle sustainability and availability of cash in dividend distribution. In fact, research shows that firms, even if characterized by positive profit, may not be able to pay dividends, given that profit is statistically higher than FCFE.

3.4. ICRS COMPARISON

A financial sustainability evaluation of the cost of debt needs ICRs application, particularly in preventing financial crisis. Again, banks could usefully apply ICRs to assess the creditworthiness of firms in the sector, even in the current state of reduced bank lending. Specifically, financial flow analysis shows that tomato processing firms considered in the sample may have difficulty paying interest on financial debt. Income margins, in fact, approximate liquidity creation where financial margins directly express the financial amount available to pay the cost of debt before distribution of dividends. It is then necessary to calculate ICRs applying both, an income approach (ICR1, which is EBITDA-based, and ICR2, which is EBIT-based) and a financial approach (ICR3, ICR4, and ICR5, which are CF, OCF, and UFCF-based, respectively). ICRs with an income approach, taking EBITDA and EBIT as the numerator (ICR1 and ICR2), have average values of 2.77 and 2.06, respectively. ICRs calculated with a financial approach (CF, OCF, and UFCF-based) are ICR3, ICR4, and ICR5 with average values 2.32, 2.01 and 0.93, respectively. The comparison of the significant differences between ICRs, calculated with both income and financial approaches, using the Wilcoxon statistic (Wilcoxon Matched-Pairs Signed-Ranks Test) for paired samples (Table Nº 7), rejects the null hypothesis of equality between mean with a two-sided test with significance at 1.00% (with the exception of couples' 1 and 5 comparison). Only for the comparisons between ICR1 and ICR3 and between ICR2 and ICR4 does the analysis show equality between mean values, expressing that an EBITDA-based ICR could correctly approximate a CF-based ICR and an EBITbased ICR could correctly approximate an OCF-based ICR.

Table 7
Comparison of ICRs with income and financial approach nonparametric approach for paired samples (T-Wilcoxon).

ICRs analysis shows that sustainability assessment applying an income-based ICRs approach could give incorrect results, even if these ICRs are the most frequently applied by firms and banks. In particular, no one incomebased ICR could correctly approximate a UFCF-based ICR, and in tomato processing firms this latest approach is particularly useful given the FA investment capital absorption, as highlighted by the cash flow statement analysis. The analysis then shows that it could be preferable to apply financial-based ICRs, particularly for tomato processing firms, to correctly express the firm's capacity to pay the cost of debt considering the financial resources actually available, even considering investment in FA.

3.5. REGRESSION ANALYSIS

Given the results of income and financial margins, the analysis will therefore identify determinants of income (PROFIT) and cash flow (FCFE) margins available for equity holders. We have applied a regression analysis to quantify the causal relationship between a variable to be explained (the dependent variable) and a set of explanatory variables (independent variables); the analysis has the aim of showing which variables are the determinants' income and financial flows. We firstly analyze relations between financial return on equity capital for a given period t (FOEt), and given independent variables. To achieve this goal we consider the explanatory capacity of a linear regression model (financial model) as expressed in equation ⁽¹⁷⁾, where FOEt is the financial return available for equity holders, as an independent variable for a given time (t). The idea underlying the financial model is to explain actual FOE (at a given time, t) with a set of explanatory variables that express capital intensity (TO, SIZE), working capital cycle duration (INV_DAYS, AR_DAYS, AP_DAYS), debt level (DER, NFP_E), operative profitability (ROS), actual income margins (EBIT, EBITDA, and PROFIT), and their respective values considered at t-1 (EBITt"1, EBITDAt"1, and PROFITt"1), even considering actual financial margins (CF, OCF, and UFCF) and their respective values considered at t-1 (CFt, OCFt, and UFCFt). In the financial model, the constant term is a, TO is turnover (VP / invested capital), INV_DAYS is the duration in days of the cycle of the inventories in stock, AR_DAYS is the duration in days of the average extension to customers, AP_DAYS is the duration in days of the average extension from suppliers, DER is debt-to-equity ratio (D/E), NFP_E is net financial position to equity ratio (NFP/E), and SIZE is the amount of capital invested in euro (total assets). The model then considers EBITDA, EBIT, and PROFIT as explanatory variables, considered in values for years t and t- 1 (EBITDAt and EBITDAt"1, EBITt and EBITt"1, and PROFITt and PROFITt"1). Financial margins CF, OCF, and UFCF are considered explanatory variables and taking in account their values for years t and t-1; thus, there are six explanatory variables (CFt and CFt"1, OCFt and OCFt"1, and UFCFt and UFCFt"1). The model is:

The financial model, as expressed in regression equation ⁽¹⁷⁾, is analyzed in Table Nº 8 and assumes a significant statistical capacity to explain FOEt values. The F statistic has high significance (F = 0.000). R2 is 0.965, while adjusted R2 has a value of 0.962. The model is then useful in explaining the greater part of FOEt variability. The DW statistic is 2.052, and the majority of the variables are significant. First, TO has a positive effect on FOE values, expressing that an increase in turnover -that is a decrease in the capital-intensive structure of assets- increases the FCFE value. Explanatory variables of FOE generation are values that express working capital (WC) cycle durations. Particularly, INV_DAYS and AR_DAYS have a negative sign, where an increase in WC durations has a negative effect on the FOE result, with a decrease available cash flow. AP_DAYS has a positive sign on the FOE, expressing an opposite situation. DER has a positive sign on the FOE, given that debt increasing generates cash, while an increase in financial debt (NFP_E) has an effect on increasing interest charges, thus reducing cash flow available. Income and financial margins at a given time, t, have an effect on FOE at the same time, t (particularly PROFITt and UFCFt). Income and financial margins at t-1 have a limited effect on FOE, with the exception of variables PROFITt"1 and UFCFt"1 with a relation significant only at the 0.05 level (twotailed).

Table 8
Extract of the multiple regression model that shows the impact on FOEt of independent variables - Financial model, equation (17).

The second considered regression model (income model) analyzes the relation between the return on equity capital for a given period, t (ROEt); that is, the amount of PROFIT available for equity holders, and a set of independent variables, the same considered in the financial regression model, as expressed in the methodological part of the article. In the income regression model, the constant term is a, TO is turnover (VP / invested capital), INV_DAYS is the duration in days of the cycle of the inventories in stock, AR_DAYS is the duration in days of the average extension to customers, AP_DAYS is the duration in days of the average extension from suppliers, DER is debt-to-equity ratio (D/E), NFP_E is net financial position to equity ratio (NFP/E), and SIZE is the amount of the capital invested in euro (total assets). The model then considers EBITDA and EBIT as explanatory variables, considered in values for the years t and t-1 (EBITDAt and EBITDAt"1, EBITt). PROFIT is not considered as an explanatory variable, as it is part of the ROE equation. Financial margins CF, OCF, UFCF, and FCFE are considered explanatory variables in their values for years t and t-1, giving then another eight explanatory variables (CFt and CFt"1, OCFt and OCFt"1, UFCFt and UFCFt"1, and FCFEt and FCEFt"1). In the income model, the set of explanatory variables is the same as in equation (17), with the exception of FCFE instead of PROFIT. We express the income model as follows:

The income regression model as expressed in equation ⁽¹⁸⁾ has a significant statistical capacity to explain FOEt values, as exposed in Table Nº 9. The F statistic has high significance (F = 0.000);R2 is 0.885, while adjusted R2 has a value of 0.882; and statistic DW is 2.121. The income model is even able to explain a large part of the variability of ROEt although with less magnitude than the financial model. TO has a positive effect on ROE values (as in SIZE), as that increase in turnover has a positive effect on the PROFIT value. ROE generation is even affected by values expressing the duration of the working capital (WC) cycle: INV_DAYS and AR_DAYS have a positive sign, expressing that an increase in WC durations has a positive effect on the ROE result, even if this causes a decrease in cash available.

Table 9
Extract of the multiple regression model that shows the impact on ROEt of independent variables - Second model, equation (18).

AP_DAYS is not even as statistically significant DER and NFP_E are.

Obviously, ROS is particularly important in increasing the ROE value, just as EBIT and EBITDA are at a given time t. Income and financial margins at a given time t also have an effect on ROE (particularly EBITDAt and EBITt and even UFCFt and FCFEt), while in the financial model, income and financial margins at t-1 time (with the exception of EBITt"1 and FCFEt"1) have a limited effect on ROE.

4. CONCLUSIONS

Analysis shows that tomato processing firms have characteristics of production and investment that affect capital structure and need to assess business cycle sustainability. The research data highlights that tomato processing firms require large amounts of capital to finance FA investments, such as buildings, plants, and equipment for tomato processing and NWC, particularly for inventories, including finished goods, and accounts receivable. Sample data gives relevant correlations between income and financial margins, even if it is possible to note statistically significant differences, particularly between economic margins, UFCF and FCFE. The analysis then highlights that profit margins could not be applied to correctly approximate financial margins as is frequently done in bank analyses and firms' evaluation. About this topic, it is necessary to note the importance to purpose financial statements that, despite its importance, is not available in 36 of the 50 firms in the sample. Financial data analysis shows that sample firms often have difficulty in paying interest charges and in distributing dividends as expressed by UFCF and FCFE values, respectively, and this is true even in the case of positive profit. About this topic, research data conduct a sustainability evaluation by applying traditional ICRs, calculated with an income approach (ICR1, ICR2); the analysis shows that income ICRs have higher values than financial ICRs (ICR3, ICR4, ICR5) and then ICR1, ICR2 could overestimate firms' capacities to pay interest charges. Financial ICRs could then be applied with higher significance, expressing more correctly the firm's capacity to pay the cost of debt. ICRs -as tested in the research- may give better information to the firm's owner to assess the management cycle's sustainability, and could even be usefully applied by credit institutions that could assess more accurately creditworthiness of firms. At the end, the research even highlights that tomato firms' management could assume wrong decisions if they only consider economic data. In fact, research highlights that tomato firms, even if characterized by positive profit, may not be able to pay dividends. A multiple regression model approach was applied, suggesting a useful approach to measure tomato processing firms' management that could be applied to other agrofood sectors, particularly if characterized by high capital intensity, then considering long term return of equity capital. The obtained results could then be applied by policy makers, operating with aid policies in favor of the tomato sector, even in the case of subsidiary guarantees issued by credit unions to improve the sustainability of the business cycle. In fact, a better understanding of the financial sustainability of a public-aided firm could improve efficient use of collective resources even considering the probability of default.

REFERENCES

Altman, E. I. (1968). Financial ratios, discriminant analysis and prediction of corporate bankruptcy. Journal of Finance, 23(4), 589-609.

Baños-Caballero, S., García-Teruel, P.J. & Martínez-Solano, P. (2014). Working capital management, corporate performance, and financial constraints. Source of the Document Journal of Business Research, 67(3), 332-338.

Barnes, P. (1982). The analysis and use of financial ratios: a Review Article. Journal of Business Finance and Accounting, 14(4), 449- 61.

Beaver, W. H. (1966). Financial Ratios as predictors of failure, empirical research in accounting. Selected Studies (1966). Supplement to Journal of Accounting Research, 4(3), 71-111.

Bodenhorn, D. (1964). A cash-flow concept of profit., Journal of Finance 19(1), 16-31.

Bonazzi, G., Iotti, M. & Paduano, F. (2012). Valutazioni di convenienza e di sostenibilità per le imprese del comparto del Prosciutto di Parma DOP: un'analisi attraverso l'applicazione di incidi economici e finanziari. Rivista di Economia Agraria, LXVII(2), 61-98.

Bonazzi, G. & Iotti, M., (2014). Interest coverage ratios (ICRs) and financial sustainability: Application to firms with bovine dairy livestockAm. J. Agric. Biological Sci, (9), 482-489.

Bond, D., Bugeja, M. & Czernkowski, R. (2012). Did Australian firms choose to switch to reporting operating cash flows using the indirect method? Australian Accounting Review, 22(1), 18-24.

Boschi, M., Girardi, A. & Ventura, M. (2014). Partial credit guarantees and SMEs financing. Journal of Financial Stability15, 182-194.

Bowen, R.M. & Owen, O. (1986). Evidence on the relationships between various earnings measures of cash flow. Accounting Review, 4(61), 713-25.

Bradbury, M. (2011). Direct or indirect cash flow statements?, Australian Accounting Review 21(2), 124-130.

Charitou, A. & Panagitodes, G. (1999). Financial analysis, future earnings and cash flows and the prediction of stocks returns: evidence for the UK, Accounting and Business Research, 29-41.

Cheng, C. S. A., Liu, C. & Schaefer, T. (1996). Earnings permanence and the incremental information content of cash flow from operations. Journal of Accounting Research, (34), 173-181.

Clacher, I., De Ricquebourg, A. D. & Hodgson, A. (2013). The value relevance of direct cash flows under International Financial Reporting Standards. Abacus, 49, 367-395.

Cleary, S. (1999). The relationship between firm Investment and financial status., Journal of Finance 54, 673-92.

Darun, M. R., Roudaki, J. & Radford, J. (2015). Forces shaping working capital management practices: A preliminary study. International Business Management, 9(3), 301-318.

De Miguel, A. & Pindado, J. (2001). Determinants of capital structure: New evidence from Spanish panel data. Journal of Corporate Finance, 7, 77-99.

Dechow, P. M. (1994). Accounting earnings and cash flow as measures of firm performance: the role of accounting accruals. Journal of Accounting and Economics, 18, 3-42.

Dechow, P. M. & Dichev, L. (2002). The quality of accruals and earnings, the role of accruals estimation errors. Accounting Review87, 35-59.

Dothan, M. (2006). Costs of financial distress and interest coverage ratios. The Journal of Financial Research, 29, 147-62.

Dunn, P. & Cheatham, L. (1993). Fundamentals of small business financial management for start-up, survival, growth, and changing economic circumstances. Managerial Finance, 19, 1-13.

European Union, EU. (1978). Directive 78/660/ EEC IV of July 25th. Brussels: European Commission, Official Journal of the European Communities, 14.8.78, No L 22/ 11. Recovered from http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:31978L0660&from=ES

Ezzamel, M., Brodie, J. & Mar-Molinero, C. (1987). Financial patterns of UK manufacturing companies. Journal of Business Finance & Accounting, 14(4), 519-536.

Farshadfar, S. & Monem, R. (2013). Further evidence on the usefulness of direct method cash flow components for forecasting future cash flows. International Journal of Accounting, 48(1), 111-133.

Fazzari, S. & Petersen, B. (1993). Working capital and fixed investment: New evidence on financing constraints. Rand Journal of Economics, (23), 328-342.

Finger, C.A. (1994). The ability of earnings to predict future earnings and cash flows., Journal of Accounting Research 5, 210-23.

Glancey, K. (1998). Determinants of growth and profitability in small entrepreneurial firms. International Journal of Entrepreneurial Behaviour & Research, 4, 18-27.

Goldstein, R., Ju, N. & Leland, H. (2001). An EBIT-based model of dynamic capital structure. Journal of Business, 74, 483-512.

Gombola, M. J., Haskin, J. E., Ketz, J. E. & Williams, D. D. (1987). Cash flow in bankruptcy prediction. Financial Management, 12, 55-65.

Grablowsky, B. J. (1984). Financial management of inventory. Journal of Small Business Management, 22, 59-65.

Gray, S., Mirkovic, A. & Ragunathan, V. (2006). The determinants of credit ratings: Australian evidence. Australian Journal of Management, 31, 333-354.

Grenberg, R.R., Johnson, G.L. & Ramesh, K. (1986). Earnings versus Cash Flow as a Predictor of Future Cash Flow Measures. Journal of Accounting, 4, 266-277.

Healy, P. M. (1985). The effect of bonus scheme on accounting decisions., Journal of Accounting and Economics 7, 85-107.

Hill, M. D., Kelly, W. G. & Highfield, M. J. (2010). Net operating working capital behavior: A first look., Financial Management 2, 783-805.

Howorth, C. & Westhead, P. (2003). The focus of working capital management in UK small firms. Management Accounting Research, 14, 94-111.

Hussain, S. & Al-Attar, A. (2003). Corporate data and future cash flows. Journal of Business Finance and Accounting, 6, 101-119.

Iotti, M. & Bonazzi, G. (2014). Life cycle flow analysis: Methodological improvement and application to the Parma PDO Ham Sector. Quality Access to Success, 15(143), 98-103.

Iotti, M. & Bonazzi, G. (2015). Application of ICRs with a net financial position (NFP) repayment approach in the Parma PDO ham sector. Journal of Food, Agriculture and Environment, 13(1), 109-114.

Italian National Institute of Statistics, ISTAT. (2014). Annuario statistico italiano 2013. Roma: ISTAT.

Kieschnick, R., La Plante, M. & Mousawwi, R. (2008). Working capital management, cor porate governance and firm value. Austin: University of North Texas University Press.

Kwon, Y. K. (1989). Accrual versus cash-basis accounting method: an agency-theoretic comparison. Journal of Accountingand Public Policy, 8, 267-281.

Lagerkvist, C. J. & Andersson, H. (1996). Taxes, inflation and financing - The rate of return to capital for the agricultural firm. European Review of Agricultural Economics, 23, 437-454.

Leland, H. (1994). Corporate debt value, bond covenants, and optimal capital structure., Journal of Finance (49), 1213-1252.

Leland, H. (1998). Agency costs, risk management, and capital structure., Journal of Finance 53, 1213-1243.

Lewellen, J. W. (2004). Predicting returns with financial ratios. Journal of Financial Economics, 74, 209-235.

Lorek, K.S. & Willinger, G. (1996). A multivariate times series prediction model for cash flow data. The, Accounting Review 11, 81-101.

Love, I., Preeve, L. & Sarria-Allende, V. (2007). Trade credit and bank credit: Evidence from recent financial crisis., Journal of Financial Economics 83, 453-69.

Moir, L. & Sudarsanam, S. (2007). Determinants of financial covenants and pricing of debt in private debt contracts: The UK evidence., Accounting and Business Research 37(2), 151-166.

Molina, C. & Preeve, L. (2009). Trade receivable policy of distressed firms and its effects on the cost of financial distress., Financial Management 38, 663-686.

Moro Visconti, R. (1999). Il cash flow return on investment CF ROI. Impresa Commerciale Industriale, 21, 34-51.

Ohlson, J. (1980). Financial Ratios and the probabilistic prediction of bankruptcy., Journal of Accounting Research 4, 109-131.

Padachi, K. (2006). Trends in working capital management and its impact on firms' performance: An analysis of Mauritian small manufacturing firms. International Review of Business Research Papers, 2, 45-58.

Peel, M. J. & Wilson, N. (1996). Working capital and financial management practices in the small firm sector. International Small Business Journal, 14, 52-68.

Rayburn, J. (1986). The association of operating cash flow and accruals with security returns., Journal of Accounting Research 7, 112-133.

Rosner, R. L. (2003). Earnings manipulation in failing firms. Contemporary Accounting Research, 2, 361-408.

Russel, P. B. (2009). The cash flow implication of managing working capital and capital investment. Journal of Business& Economic Studies, 15, 98-108.

Sgroi, F., Di Trapani, A.M., Testa, R. & Tudisca, S. (2014). Economic sustainability of early potato production in the Mediterranean area. Am. J. Applied Sci., 11, 1598-1603.

Sloan, R. (1996). Do stock prices fully reflect information in accrual and cash flows about future earnings. The, Accounting Review 9, 71-93.

Taylor, D. (2011). Optimizing working capital. Business Credit, 6, 12-15.

Testa, R., Di Trapani, A.M., Sgroi, F. & Tudisca, S. (2014). Economic sustainability of Italian greenhouse cherry tomato. Sustainability (Switzerland), 6(11), 7967-7981.

Wang, Z. & Eichenseher, J. (1998). Informativeness and predictability of cash flows. Journal of Applied Business Research, 13, 21-32.

Weinraub, H. J. & Visscher, S. (1998). Industry practice relating to aggressive and conservative working capital policies. Journal of Financial and Strategic Decisions, 11, 21- 37.

Wilson, G. P. (1987). The incremental information content of accrual and funds components of earnings after controlling for earnings. The, Accounting Review 52, 293- 322.

World Processing Tomato Council, WPTC (2014). World production estimate as of 19 February 2014. World Processing Tomato Council Releases. Sorgues (France): WPTC.

Notes