石化企业利润的波动主要来自于原料端
石化企业利润的波动主要来自于原料端
longpu2017
为聚烯烃创造着，快乐着，骄傲着！
以下文章来源于尚能说化 PetroChemTalks ，作者尚能说化
用我们的方式去观察和呈现能化产业与宏观经济。OUR way to observe and present the PetroChem industry and macroeconomics.
好久不见的朋友们，记得加星。你加星！
Friends, long time no see. Please add a star to follow.
Full English version is at the end.
如题所示，一个典型的以石脑油为原料的石化企业，它的加工利润波动或变化的重点是来自于原料端--石脑油，而非产品端--聚乙烯聚丙烯，甚至某些时候是来自于汇率端 -- 人民币美金的汇率。
一、加工利润公式和加工利润波动的公式不一样
-
加工利润 = ((LLDPE*2+PP)-((3*Nap*1.35+410)*Rate*1.17+450))/3
-
△ 加工利润 = -1010* △ Rate+0.66* △ LLD+0.34* △ PP-10* △ Naphtha
举个例子, Y=aX1+bX2 这是二元一次公式，自变量X1和X2的系数分别是a和b。
理论而言，△Y=a△X1+b△X2
实际而言，如果我们用Y，X1和X2的“实际数据”进行多元一次回归，得到的公式 △Y=a1△X1+b2△X2。往往a1≠a b2≠b，甚至有时差别很大。
这个原因是：自变量的波动对于因变量波动的关系 ≠ 自变量的绝对值对于因变量的绝对值的关系。譬如一个极端例子：Y=X1+2X2，如果在“实际数据”中，X2基本不变化，也就是△X2 --> 0，那么 △Y=△X1+0，Y的波动只与X1的波动有关。
所以，回到石脑油的生产加工利润，在 《加工利润的套保工具，帮助石化企业对冲了真实利润吗？(2) Hedge on Profit of Petrochemical》 文章中推导出的加工利润的公式和加工利润波动的公式分布如下:
首先，对加工利润，基于过去10年数据进行回归分析后，推导出来的加工利润波动或者变化的公式，两者之间是很不一样的。其次， 可以看到四个自变量波动的系数差异很大，这说明各自波动对于加工利润波动的贡献差异很大：
△Rate：-1010
△ LLDPE: +0.66
△ PP: +0.34
△ Naphtha: -10
也就是说，各个自变量 每增加1个单位（1元），汇率Rate将造成加工利润1010元的下降，LLDPE将造成0.66元的上涨，PP是0.34元，Naphtha造成 -10元的下跌。
二、自变量波动对于加工利润贡献的实际情况
自变量波动的系数大小只是在理论上说明该自变量变化的贡献大小，我们还需要观察在实际过程中，它自己波动的情况，也就是它自己变化的大小。举一个例子，△Rate的系数是-1010，非常大，但是实际过程中，它波动幅度很小，短期一般在0.1~0.2，那么Rate这个自变量波动的实际贡献是很小的。
我们用四者的标准偏差St. Dev来大致表征一下四者的波动情况（过去2017年至今）：
Rate: 0.24
LLDPE: 958 RMB
PP: 815 RMB
Naphtha: 150 USD
但是这并不能说明各自对于加工利润波动贡献的大小程度，因为这还和四者各自变化的持续性有关。 举个例子：假如LLDPE的价格是上下波动11000、10000、 11000、10000、11000、10000 对利润不会有影响只是波动，但是如果价格是持续的11000、11000、11000、10000、10000、10000对利润的影响就很大。
所以，我们用实际的数据带入到公式，观察它们对于加工利润的累积贡献。我们发现石脑油对于加工利润波动的贡献是最大，其次是Rate，LLDPE和PP。
如果我们再计算一下他们各自贡献的百分比占比，我们可以看到石脑油对于加工利润的影响是处于绝对主导地位。
原料和产品端我们都很熟悉，也是平常我们关注的环节。 那么关于波动系数是1010的汇率呢？ 这么大的系数会造成什么样的影响？下图中我们可以看到，它的波动累积对于加工利润波动的贡献可以达到+50%~-100%， 还是一个相对显著的因素。
如果我们再详细观察某些阶段：2020/01~2021/04 和 2022/03~2023/03，这是两段近年来加工利润显著扩展的时期。但是我们可以发现：1、汇率的影响很大，与产品端几乎一致。2、汇率在两个利润扩展期分别是正贡献和负贡献，对加工利润的变化形成完全不同的影响。 汇率问题在某些时期是非常值得石化企业进行关注和管理的。
三、石脑油的波动为什么那么大
我们已经发现石脑油波动对于加工利润的贡献是很大的，一方面是因为它的波动系数是10，另一方面也是源于△Naphtha本身很大，尤其是过去三年非常明显。
并不深究地分析，非常简单地假设： 对于石脑油和聚烯烃而言，下游的需求都足够分散，谁供应越集中那么造成的价格波动也就可能越大。
1、聚烯烃一共有多少套装置 ？
-
2、中国有多少个炼油厂？
中国大陆总计
PE：共149套， 平均规模22万吨/年
PP：共216套，平均规模20万吨/年
其中含外资的装置：PE：共37套 PP：共27套
约217家，平均规模430万吨/年，52万吨/年石脑油（按12%石脑油出率）
数量少于聚烯烃，规模大于聚烯烃，并且具有众多的千万吨级炼厂，如果考虑到一些小型炼厂的关停并转， 石脑油的供应可能更为集中。
在 《“原油—成品油/石脑油—化工品”谁是链上的老大？》 一文中我们谈到过两个概念石脑油既是炼厂成品油的“副产品”，也是“竞争品”，它的价格受成品油和原油的影响很大。所以结合供应端情况和成品油链条的特点， 就价格而言，石脑油相对于聚烯烃波动可能更大： △ Naphtha > △ PE/ △ PP
所以，原料端的石脑油价格是对石化企业加工利润影响最大的因素，而产品端的影响要弱化很多，同时需要注意的是，汇率的影响往往很大且相对隐蔽，我们也需要关注。
As the title suggests, the focus of fluctuations or changes in the processing margins of a typical naphtha-based petrochemical company comes from the feedstock end - naphtha, rather than the product end - polyethylene polypropylene, or even in some cases from the exchange rate end - the exchange rate of the RMBUSD.
I. The processing profit formula is not the same as the formula for processing profit fluctuations
For example, Y=aX1+bX2 This is a binary equation with coefficients a and b for the independent variables X1 and X2, respectively.
Theoretically, ∆Y=a∆X1+b∆X2
In practice, if we use the "actual data" of Y, X1 and X2 to conduct multiple primary regression, we get the formula △Y=a1△X1+b2△X2. a1≠a b2≠b, and sometimes even very different.
The reason is that the relationship between the fluctuation of the independent variable and the fluctuation of the dependent variable ≠ the relationship between the absolute value of the independent variable and the absolute value of the dependent variable. For example, in an extreme case: Y=X1+2X2, if in the "actual data", X2 is basically unchanged, i.e., △X2 --> 0, then △Y=△X1+0, and the fluctuation of Y is related to the fluctuation of X1 only.
So, back to the production and processing profit of naphtha, in the "Processing Profit Hedge Tool, Helping Petrochemical Enterprises Hedge their Real Profit? (2) Hedge on Profit of Petrochemical" article, the formula of processing profit and the formula of processing profit fluctuation derived are distributed as follows.
Processing Profit = ((LLDPE*2+PP)-((3*Nap*1.35+410)*Rate*1.17+450))/3
∆Processing Profit = -1010*∆Rate+0.66*∆LLD+0.34*∆PP-10*∆Naphtha
First, for processing profit, the formula for the fluctuation or change in processing profit derived from the regression analysis based on the data of the past 10 years is very different between the two. Second, it can be seen that the coefficients of the four independent variable fluctuations are very different, which indicates that the contribution of their respective fluctuations to the fluctuation of processing profits varies greatly:
△Rate: -1010
△LLDPE: +0.66
△PP: +0.34
△Naphtha: - 10
That is, for each unit ($1) increase in each independent variable, the exchange rate Rate will cause a $1010 decrease in processing profits, LLDPE will cause a $0.66 increase, PP is $0.34, and Naphtha causes a -10 decrease.
Second, the actual situation of the contribution of the fluctuation of the independent variable to the processing profits
The size of the coefficient of independent variable fluctuation is only a theoretical description of the size of the contribution of the independent variable changes , we also need to observe in the actual process, its own fluctuations, that is, the size of its own changes. As an example, the coefficient of △Rate is -1010, which is very large, but in the actual process, it fluctuates very little, and the short-term is usually in the range of 0.1~0.2, so the actual contribution of the fluctuation of this independent variable of Rate is very small.
So, we use the actual data to bring into the formula, we find that the contribution of naphtha to the fluctuation of processing profit is the largest, followed by Rate, LLDPE and PP.
If we then calculate the percentage share of their respective contributions, we can see that the impact of naphtha on processing margins is absolutely dominant.
The feedstock and the product side are familiar to us and are the usual segments that we focus on. So what about the volatility factor of 1010 for the exchange rate? What is the impact of such a large coefficient? In the chart below we can see that the cumulative contribution of its fluctuations to the fluctuations in processing profits can reach +50% to -100%, still a relatively significant factor.
If we look at certain phases in more detail: 2020/01~2021/04 and 2022/03~2023/03, these are two periods of significant expansion of processing profits in recent years. However, we can find that: 1. The impact of exchange rate is significant and almost the same as that of the product side. 2, the exchange rate in the two profit expansion period is a positive contribution and negative contribution, respectively, on the changes in processing profits to form a completely different impact. The exchange rate issue is very worthwhile for petrochemical companies to pay attention to and manage in some periods.
III. Why naphtha volatility is so large
We have found that the contribution of naphtha volatility to processing profit is large, partly because its volatility coefficient is 10, and partly because △Naphtha itself is large.
Not a deep analysis, very simple assumption: for naphtha and polyolefins, downstream demand is sufficiently decentralized, who supply is more concentrated then the price volatility caused by the greater possible.
1. How many polyolefin plants are there in total?
Total in mainland
PE: a total of 149 sets, the average size of 220,000 tons / year
PP: 216 sets, average size of 200,000 tons / year
Including foreign-invested installations: PE: a total of 37 sets PP: a total of 27 sets
2、How many refineries are there in China?
About 217, with an average size of 4.3 million tons/year and 520,000 tons/year of naphtha (at 12% naphtha output rate)
The number is less than polyolefin, the scale is larger than polyolefin, and there are many 10 million tons of refineries, if you take into account the closure of some small refineries, the supply of naphtha may be more concentrated.
In the article "Crude oil - refined oil products / naphtha - chemicals" who is the boss of the chain? we talked about two concepts naphtha is not only a "by-product" of refinery products, but also a "competitive product", and its price is greatly influenced by refined products. Therefore, combining the supply side situation and the characteristics of the refined oil chain, in terms of price, naphtha may fluctuate more than polyolefin: △Naphtha > △PE/△PP
Therefore, the price of naphtha at the feedstock end is the factor that has the greatest impact on the processing profit of petrochemical enterprises, while the impact of the product end is much weaker, and it should be noted that the impact of the exchange rate tends to be large and relatively hidden, and we need to pay attention to it as well.
议程更新
2023.8.16
全天报到 酒店大堂
14:30-17:30 龙朴私享会（定向邀约制）
2023.8.17
8:30-17:00 全体大会
2023.8.18
8:30-12:00 全体大会
14:00-17:00 分论坛
分会场一：POE/ α-烯烃 分会场 分会场二：EVA分会场 分会场三：COC/COP 分会场
分会场四：聚烯烃挤出造粒技术专场
冠名单位： 沈阳金锋特种设备股份有限公司
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
国泰君安
龙朴聚烯烃
Univation
陕煤研究院
利安德巴赛尔
独山子石化研究院
扬子石化研究院
齐鲁石化研究院
赛科石化
兰州石化
华北化销
东华大学
浙江工业大学
上海开维喜阀门
阿特拉斯
佳华精化
......
赛伍
盛虹石化（斯尔邦）
大毛牛
亚培烯
卫星化学
阿科力
金发科技
……邀请中
-
重磅！“三桶油”最新人事变动
-
重大调整！中化泉州100万吨/年乙烯项目重新公示
-
挤压造粒专家多图详解丨造粒三部件的安装、检验与调整
-
浙江民营化工巨头迎来36岁“二代”女掌门人！615亿大炼化项目正式敲定！