fdic 0.1.3

Simple tools for accessing FDIC public data.

Introduction

This project provides some tools to access data from the FDIC public API and data resources.

Work in progress; more docs coming soon.

Usage

To try out these tools, install via pip install fdic or download the source code from GitHub. Then start a python interpreter and do something like the following:

First we import various modules and set ftool to be the tool we want:

>>> import fdic.query, json
>>> ftool = fdic.query.FDICTools

Now we can get the institutional data from the FDIC and sort it first by assets and then by deposits. We print the result in JSON format as a sanity check and see the largest 3 banks by assets:

>>> inst_data = ftool.get_sorted_inst_data()
>>> top_3_by_assets = inst_data[:3]
>>> print(json.dumps({i['NAME']: {n: i[n] for n in ['ASSET', 'DEP']}
...      for i in top_3_by_assets}, indent=2))
{
  "JPMorgan Chase Bank, National Association": {
    "ASSET": "3201942000",
    "DEP": "2440722000"
  },
  "Bank of America, National Association": {
    "ASSET": "2418508000",
    "DEP": "2042255000"
  },
  "Citibank, National Association": {
    "ASSET": "1766752000",
    "DEP": "1399631000"
  }
}

Next we pull in the Uniform Bank Performance Report data. Note that you will have to download a zip file with this data first. If you don’t, you will get a NeedUBPRZipFile exception telling you how to download the necessary file.

To save time in parsing, we can provide an rssd_filter to just get data for the top 200 banks by assets. After getting the data, we sort by the UBPRE569 field (unrealized losses as a percent of tier 1 capital for held-to-maturity assets) and then print the data:

>>> ubpr_data = ftool.get_ubpr_inst_data(rssd_filter={
...     i['FED_RSSD'] for i in inst_data[:50]})
>>> htm_data = list(sorted(ubpr_data, key=lambda i: i['UBPRE569']))
>>> print(json.dumps({i['NAME']: {n: i[n] for n in ['NAME', 'UBPRE569']}
...      for i in htm_data[:3]}, indent=2))
{
  "Silicon Valley Bank": {
    "NAME": "Silicon Valley Bank",
    "UBPRE569": -89.2
  },
  "Bank of America, National Association": {
    "NAME": "Bank of America, National Association",
    "UBPRE569": -59.95
  },
  "Charles Schwab Bank, SSB": {
    "NAME": "Charles Schwab Bank, SSB",
    "UBPRE569": -46.87
  }
}

You can find a list of the UBPR codes at the Federal Reserve. For example if you are interested in both UBPRE569 (unrealized losses on the held-to-maturity portfolio as percent of tier 1 capital) as well as UBPRM037 (appreciation in available for sale securities / percent of available for sale securities), you could do something like the following:

>>> codes = {
...    'UBPRE569': {'convert': float},
...    'UBPRM037': {'convert': float},
... }
>>> ubpr_data = ftool.get_ubpr_inst_data(rssd_filter={
...     i['FED_RSSD'] for i in inst_data[:50]}, codes=codes)
>>> htm_data = list(sorted(ubpr_data, key=lambda i: i['UBPRE569']))
>>> print(json.dumps({i['NAME']: {n: i[n] for n in (['NAME']+list(codes))}
...      for i in htm_data[:3]}, indent=2))
{
  "Silicon Valley Bank": {
    "NAME": "Silicon Valley Bank",
    "UBPRE569": -89.2,
    "UBPRM037": -8.86
  },
  "Bank of America, National Association": {
    "NAME": "Bank of America, National Association",
    "UBPRE569": -59.95,
    "UBPRM037": -2.0
  },
  "Charles Schwab Bank, SSB": {
    "NAME": "Charles Schwab Bank, SSB",
    "UBPRE569": -46.87,
    "UBPRM037": -8.17
  }
}