【正文】 產(chǎn)作為獨(dú)立變量，我們選擇了幾個(gè)實(shí)證文獻(xiàn)中出現(xiàn)過的。 分析方法 本研究利用多元回歸分析，以檢驗(yàn)上述假設(shè)制定。 分析過程如下三個(gè)階段。 4 結(jié)果與討論 統(tǒng)計(jì)性描述 表 1 報(bào)告的 是 解釋變量的描述性統(tǒng)計(jì)。平均短期負(fù)債比率約為 ％，而長(zhǎng)期債務(wù)比率僅為 ％。要看到中小企業(yè)資本結(jié)構(gòu)的變化，有必要考慮企業(yè)所有權(quán)的杠桿作用。 t 檢驗(yàn)分析 結(jié)果顯示，有 顯著 差異的杠桿水平為 的國(guó)有和民營(yíng)中小企業(yè)之間的所有措施。為什么是財(cái)務(wù)困境的理論解釋在越南的中小型企業(yè)的風(fēng)險(xiǎn)和資本結(jié)構(gòu)之間的關(guān)聯(lián)沒有什么用處呢？在 19982020 年期間，信貸市場(chǎng)仍處于管制狀態(tài)，利率由越南國(guó)家銀行 來設(shè)定 ，而不是由市場(chǎng)力量。我們的結(jié)果是相似的與黃宋（ 2020）在中國(guó)的調(diào)查結(jié)果報(bào)告。這意味著，作為其營(yíng)運(yùn)資金的方式上，公司的規(guī)模有很強(qiáng)的影響力。 在 研究中小企業(yè)的資本結(jié)構(gòu)決定因素時(shí) ， 企業(yè)所有權(quán)被認(rèn)為是一種特殊的因素。換句話說，國(guó)有中小型企業(yè)比民營(yíng)中小企業(yè)更容易獲得銀行貸款。這些公司的所有者權(quán)益中除未分配利潤(rùn)，來自國(guó)家預(yù)算。到 2020 年，國(guó)有中小企業(yè)一般有訪問權(quán)限的國(guó)有商業(yè)銀行的抵押物，利率，擔(dān)?；虻谌降男庞妙~度。短期負(fù)債占資本結(jié)構(gòu)的一個(gè) 顯著 比例，而 在越南 中小型企業(yè)很少采用長(zhǎng)期債務(wù)。與銀行的關(guān)系 越親近， 中小企業(yè)獲得金額較大的銀行貸款的 可能性越大 。也揭示了一個(gè)積極的業(yè)務(wù)風(fēng)險(xiǎn)和債務(wù)比率之間的關(guān)系，政府應(yīng)放寬利率，其目的不僅是創(chuàng)造一個(gè)安全的銀行系統(tǒng)，也迫使中小企業(yè)設(shè)立融資結(jié)構(gòu)，是適合他們的具體業(yè)務(wù)的風(fēng)險(xiǎn)程度。因此，民營(yíng)中小企業(yè)的管理者應(yīng)該意識(shí)到重要性的披露以及財(cái)務(wù)報(bào)表 的 編制，與銀行建立的信任度， 以達(dá)到 增加透明度的目的。在未來，公司的財(cái)務(wù)數(shù)據(jù)變得更加可靠和容易獲得的，以后可以用較長(zhǎng)的時(shí)間進(jìn)行 研究，以考察越南中小企業(yè)的資本結(jié)構(gòu)在經(jīng)濟(jì)周期的不同階段。 Miller 1977), firms have to trade off between the costs of financial distress, agency costs (Jensen and Meckling 1976) and tax benefits, so as to have an optimal capital structure. However, asymmetric information and the pecking order theory (Myers and Majluf 1984。 Pandey 2020). ? Size = Natural logarithm of the number of employees (Heshmati 2020). Methods of Analysis This study utilized multiple regression analysis to test the hypotheses formulated above. We calculated fouryear mean values of dependent and independent variables, except for the cases of firm growth and profitability. For determinants related to managers’ behaviour, we also employed factor analysis, through the Principal Component technique, Next, the factor scores are estimated and utilized for further multivariate analysis. Since the factor scores are generated through an orthogonal transformation, they do not pose any multicollinearity problems for the regression equation. The analysis process follows three stages. In the first stage, we conduct regressions of all determinants related to a firm’s characteristics(growth, business risk and size) on various measures of capital structure. In the second stage, we add a dummy variable to consider the effect of firm ownership on capital structure. In the last stage, we examine the influence of all determinants on capital structure. 13 4 Result and discussion Descriptive Statistics Table 1 reports the descriptive statistics of dependent and explanatory variables. Over the period ， SMEs in Vietnam had an average debt ratio about per cent. However, the debt ratio variation across the sample was large, ranging from a maximum debt ratio of per cent and a minimum of per cent. With respect to debt maturity, we find that most SMEs employ more shortterm liabilities than longterm debt to finance their operations. The average shortterm liabilities ratio was approximately per cent, and the longterm debt ratio was just per cent. Shortterm liabilities employed by SMEs are highly varied, such as mercial bank loans, trade credits from suppliers, advance payments from clients, borrowing from friends or relatives, and some other sources. The other shortterm liabilities ratio, representing mostly financing from works, account for a relatively high proportion ( per cent) of total assets. Clearly,financial sources from works play a relatively important role in SMEs’ capital structure. To see the variation of capital structure in SMEs, it is necessary to consider leverage by firm ownership. Stateowned SMEs have more debt than private SMEs, with debt ratios of per cent and per cent, respectively. Among private SMEs, jointstock panies employ more debt than limited or proprietary firms, the debt ratios being per cent, per cent and per cent, respectively. The differences in capital structure are also found in other measures of leverage. The results of ttest analysis show that there are significant differences in 14 all measures of leverage between stateowned and private SMEs at level. This result supports the notion of ownership structure as a determinant of a firm’s capital structure. Empirical Analysis and Result Discussion Our results also indicate that SMEs with higher operating risk tend to use more debt in general, and shortterm liabilities in particular. These findings do not support hypothesis 3 and conflict with theory of financial distress. Why is the theory of financial distress not useful in explaining the association between risk and capital structure in Vietnam’s SMEs? During the period 1998–2020, the credit market was still regulated, and interest rates were contained within a band set by the State Bank of Vietnam, rather than by market forces. Commercial banks were only allowed to offer interest rates within the confines of that band. As a result, panies with high business risk could still get bank loans at interest rates that were lower than if interest rates were not fixed by the central bank. This is the main reason explaining why Vietnamese SMEs with high risk can also maintain a high debt ratio. Our results are similar with findings reported by Huang and Song (2020) in China. A mechanism of controlled interest rates will negati