交通拥堵是城市发展中出现的集聚不经济现象，对经济社会、环境、身心健康都会造成多方面影响。目前，对于交通拥堵经济成本的领域研究尚不充分，不同城市由于计量方式的不同，推算的经济成本无法进行直接比较，各研究的统计年份也并不连续。因此，本文首先通过对交通延误成本、燃油过量消耗成本两类拥堵经济成本进行测算，计算了中国87个各级城市的交通拥堵经济成本以及拥堵福利损失率（交通拥堵经济成本占人均GDP比值），得出了两者在2016-2018年、2022年的区间分布结构与变动趋势，并对全国范围内的交通拥堵产生的经济影响进行研判。结论表明，第一，研究区域内，2016-2018年期间，超过一半的城市拥堵经济成本属于上升状态，并且始终集中在2,500-5,000元/人的中等成本区间。大部分城市的拥堵福利损失率变化趋势呈现下降趋势，并集中在2.5%-7.5%的中高损失区间。对比疫情前，2022年城市拥堵经济成本和拥堵福利损失率均下降明显，但北京与上海的拥堵经济影响依旧十分严重。第二，城市交通拥堵的平均经济成本随着城市等级的提升而加剧，并且除一线城市外，各线城市均在2016-2018年期间表现为上升趋势。平均福利损失率随着城市等级的提升呈现U型特征，二线城市的平均福利损失率最低，各线城市均在2016-2018年期间表现为下降趋势。

随后，为了解释为什么领域内研究对多中心发展是否降低了拥堵程度众说纷纭，本文引入非通勤因素用于研究交通拥堵。通过运用Landscan人口栅格数据集、高德城市交通分析报告与百度POI大数据，研究交通拥堵与中国85个多中心城市主副中心间服务设施分布差异的关系。结果表明，第一，娱乐购物类服务设施随着中心间分布均匀度上升而加重城市交通拥堵程度，公园景点类服务设施随着中心间分布均匀度上升而缓解城市交通拥堵程度。通勤距离的增加显著加剧交通拥堵程度，但影响有限，平均每增加1公里的单程通勤距离，OCDI（Overall Congestion Delay Index）将上升0.018。第二，与紧凑发展相关的紧凑度指标与人口密度都显著影响交通拥堵。紧凑度指标描写城市形状紧凑程度，城市形状越接近圆形，越减轻交通拥堵程度；人口密度越高，交通拥堵程度越高，每上升1%，OCDI将相应增加0.162。

本文认为随着经济活力的不断回升，降低交通拥堵的经济影响迫在眉睫，可以从降低交通拥堵程度、降低车均行驶里程、提高燃油使用效率三个角度减弱拥堵的经济影响。本文认为城市应该推动休闲娱乐业的集聚化发展，促进绿化和旅游资源均衡分布，正确权衡紧凑发展产生的集聚经济和因此造成的拥堵损失，以降低交通拥堵程度。另外，尽可能降低通勤距离，建立具有就业吸附力的副中心，避免建立单纯的居住型副中心。同时，加快公共交通建设，提高公共交通服务水平，推进绿色出行，并积极推广新能源汽车的应用以减少对传统燃油的依赖、提高燃油使用效率。

Traffic congestion is an uneconomical phenomenon that occurs in urban development, causing various impacts on the economy, society, environment, and physical and mental health. This article first calculates two types of congestion economic costs, namely traffic delay cost and fuel overconsumption cost, and calculates the traffic congestion economic cost and congestion welfare loss rate (the ratio of economic cost of traffic congestion to per capita GDP) of 87 cities at all levels in China. The interval distribution structure and change trend of the two in 2016-2018 and 2022 are obtained, and the economic impact of traffic congestion on a national scale is analyzed. The conclusion indicates that, firstly, during the period of 2016-2018, more than half of the economic costs of urban congestion in the study area were on the rise and remained concentrated in the medium cost range of 2500-5000 yuan/person. The trend of congestion welfare loss rate in most cities is decreasing, concentrated in the medium to high loss range of 2.5% -7.5%. Compared to before the epidemic, the economic cost and welfare loss rate of urban congestion in 2022 have significantly decreased, but the economic impact of congestion in Beijing and Shanghai is still very serious. Secondly, the average economic cost of urban traffic congestion intensifies with the increase of city level, and except for first tier cities, all cities showed an upward trend from 2016 to 2018. The average welfare loss rate shows a U-shaped pattern as the city level increases, with second tier cities having the lowest average welfare loss rate, and all tiers of cities showing a downward trend from 2016 to 2018.

Subsequently, this article studied the influencing factors of traffic congestion. By using the Landscan population grid dataset, Gaode urban traffic analysis report, and Baidu POI big data, this study investigates the relationship between traffic congestion and the distribution differences of service facilities between the main and sub centers of 85 multi center cities in China. The results indicate that firstly, leisure and entertainment service facilities exacerbate urban traffic congestion as the distribution uniformity between centers increases, while travel and tourism service facilities alleviate urban traffic congestion as the distribution uniformity between centers increases. The increase in commuting distance significantly exacerbates traffic congestion, but the impact is limited. On average, for every 1 kilometer increase in one-way commuting distance, OCDI will increase by 0.018. Secondly, both compactness indicators and population density related to compact development significantly affect traffic congestion. The compactness index describes the compactness of a city's shape, where the closer the city's shape is to a circular shape, the less traffic congestion is reduced; The higher the population density, the higher the degree of traffic congestion. For every 1% increase, OCDI will correspondingly increase by 0.162.

This article believes that with the continuous recovery of economic vitality, reducing the economic impact of traffic congestion is urgent. The economic impact of congestion can be weakened from three perspectives: reducing the degree of traffic congestion, reducing the average driving distance of vehicles, and improving fuel efficiency. This article believes that we should promote the agglomeration development of the leisure and entertainment industry, promote the balanced distribution of green and tourism resources, and correctly balance the agglomeration economy generated by compact development and the resulting congestion losses, in order to reduce the degree of traffic congestion. In addition, try to reduce commuting distance as much as possible, establish sub centers with employment attraction, and avoid establishing simple residential sub centers. At the same time, we will accelerate the construction of public transportation, improve the level of public transportation services, promote green travel, and actively promote the application of new energy vehicles to reduce the average driving distance and improve fuel efficiency.